CINXE.COM

<!DOCTYPE html> <html class="client-nojs vector-feature-language-in-header-enabled vector-feature-language-in-main-page-header-disabled vector-feature-sticky-header-disabled vector-feature-page-tools-pinned-disabled vector-feature-toc-pinned-clientpref-1 vector-feature-main-menu-pinned-disabled vector-feature-limited-width-clientpref-1 vector-feature-limited-width-content-enabled vector-feature-custom-font-size-clientpref-1 vector-feature-appearance-pinned-clientpref-1 vector-feature-night-mode-enabled skin-theme-clientpref-day vector-toc-available" lang="en" dir="ltr"> <head> <meta charset="UTF-8"> <title>Convolutional neural network - Wikipedia</title> <script>(function(){var className="client-js vector-feature-language-in-header-enabled vector-feature-language-in-main-page-header-disabled vector-feature-sticky-header-disabled vector-feature-page-tools-pinned-disabled vector-feature-toc-pinned-clientpref-1 vector-feature-main-menu-pinned-disabled vector-feature-limited-width-clientpref-1 vector-feature-limited-width-content-enabled vector-feature-custom-font-size-clientpref-1 vector-feature-appearance-pinned-clientpref-1 vector-feature-night-mode-enabled skin-theme-clientpref-day vector-toc-available";var cookie=document.cookie.match(/(?:^|; )enwikimwclientpreferences=([^;]+)/);if(cookie){cookie[1].split('%2C').forEach(function(pref){className=className.replace(new RegExp('(^| )'+pref.replace(/-clientpref-\w+$|[^\w-]+/g,'')+'-clientpref-\\w+( |$)'),'$1'+pref+'$2');});}document.documentElement.className=className;}());RLCONF={"wgBreakFrames":false,"wgSeparatorTransformTable":["",""],"wgDigitTransformTable":["",""],"wgDefaultDateFormat":"dmy", "wgMonthNames":["","January","February","March","April","May","June","July","August","September","October","November","December"],"wgRequestId":"fe4fe12a-0165-44b3-bb88-7164169f4ae6","wgCanonicalNamespace":"","wgCanonicalSpecialPageName":false,"wgNamespaceNumber":0,"wgPageName":"Convolutional_neural_network","wgTitle":"Convolutional neural network","wgCurRevisionId":1257441256,"wgRevisionId":1257441256,"wgArticleId":40409788,"wgIsArticle":true,"wgIsRedirect":false,"wgAction":"view","wgUserName":null,"wgUserGroups":["*"],"wgCategories":["CS1 maint: location missing publisher","CS1 maint: multiple names: authors list","CS1: long volume value","CS1 Japanese-language sources (ja)","Webarchive template wayback links","All articles with dead external links","Articles with dead external links from July 2022","CS1 German-language sources (de)","Articles with short description","Short description is different from Wikidata","Articles needing additional references from June 2019", "All articles needing additional references","All articles with specifically marked weasel-worded phrases","Articles with specifically marked weasel-worded phrases from April 2023","All articles with unsourced statements","Articles with unsourced statements from October 2017","Articles containing explicitly cited British English-language text","All articles needing examples","Articles needing examples from October 2017","Articles with unsourced statements from March 2024","Articles with specifically marked weasel-worded phrases from December 2018","Articles needing additional references from June 2017","Wikipedia articles needing clarification from December 2018","Articles with unsourced statements from June 2019","Neural network architectures","Computer vision","Computational neuroscience"],"wgPageViewLanguage":"en","wgPageContentLanguage":"en","wgPageContentModel":"wikitext","wgRelevantPageName":"Convolutional_neural_network","wgRelevantArticleId":40409788,"wgIsProbablyEditable":true ,"wgRelevantPageIsProbablyEditable":true,"wgRestrictionEdit":[],"wgRestrictionMove":[],"wgNoticeProject":"wikipedia","wgCiteReferencePreviewsActive":false,"wgFlaggedRevsParams":{"tags":{"status":{"levels":1}}},"wgMediaViewerOnClick":true,"wgMediaViewerEnabledByDefault":true,"wgPopupsFlags":0,"wgVisualEditor":{"pageLanguageCode":"en","pageLanguageDir":"ltr","pageVariantFallbacks":"en"},"wgMFDisplayWikibaseDescriptions":{"search":true,"watchlist":true,"tagline":false,"nearby":true},"wgWMESchemaEditAttemptStepOversample":false,"wgWMEPageLength":100000,"wgRelatedArticlesCompat":[],"wgCentralAuthMobileDomain":false,"wgEditSubmitButtonLabelPublish":true,"wgULSPosition":"interlanguage","wgULSisCompactLinksEnabled":false,"wgVector2022LanguageInHeader":true,"wgULSisLanguageSelectorEmpty":false,"wgWikibaseItemId":"Q17084460","wgCheckUserClientHintsHeadersJsApi":["brands","architecture","bitness","fullVersionList","mobile","model","platform","platformVersion"], "GEHomepageSuggestedEditsEnableTopics":true,"wgGETopicsMatchModeEnabled":false,"wgGEStructuredTaskRejectionReasonTextInputEnabled":false,"wgGELevelingUpEnabledForUser":false};RLSTATE={"ext.globalCssJs.user.styles":"ready","site.styles":"ready","user.styles":"ready","ext.globalCssJs.user":"ready","user":"ready","user.options":"loading","ext.cite.styles":"ready","ext.math.styles":"ready","skins.vector.search.codex.styles":"ready","skins.vector.styles":"ready","skins.vector.icons":"ready","jquery.makeCollapsible.styles":"ready","ext.wikimediamessages.styles":"ready","ext.visualEditor.desktopArticleTarget.noscript":"ready","ext.uls.interlanguage":"ready","wikibase.client.init":"ready","ext.wikimediaBadges":"ready"};RLPAGEMODULES=["ext.cite.ux-enhancements","mediawiki.page.media","site","mediawiki.page.ready","jquery.makeCollapsible","mediawiki.toc","skins.vector.js","ext.centralNotice.geoIP","ext.centralNotice.startUp","ext.gadget.ReferenceTooltips","ext.gadget.switcher", "ext.urlShortener.toolbar","ext.centralauth.centralautologin","mmv.bootstrap","ext.popups","ext.visualEditor.desktopArticleTarget.init","ext.visualEditor.targetLoader","ext.echo.centralauth","ext.eventLogging","ext.wikimediaEvents","ext.navigationTiming","ext.uls.interface","ext.cx.eventlogging.campaigns","ext.cx.uls.quick.actions","wikibase.client.vector-2022","ext.checkUser.clientHints","ext.growthExperiments.SuggestedEditSession","wikibase.sidebar.tracking"];</script> <script>(RLQ=window.RLQ||[]).push(function(){mw.loader.impl(function(){return["user.options@12s5i",function($,jQuery,require,module){mw.user.tokens.set({"patrolToken":"+\\","watchToken":"+\\","csrfToken":"+\\"}); }];});});</script> <link rel="stylesheet" href="/w/load.php?lang=en&modules=ext.cite.styles%7Cext.math.styles%7Cext.uls.interlanguage%7Cext.visualEditor.desktopArticleTarget.noscript%7Cext.wikimediaBadges%7Cext.wikimediamessages.styles%7Cjquery.makeCollapsible.styles%7Cskins.vector.icons%2Cstyles%7Cskins.vector.search.codex.styles%7Cwikibase.client.init&only=styles&skin=vector-2022"> <script async="" src="/w/load.php?lang=en&modules=startup&only=scripts&raw=1&skin=vector-2022"></script> <meta name="ResourceLoaderDynamicStyles" content=""> <link rel="stylesheet" href="/w/load.php?lang=en&modules=site.styles&only=styles&skin=vector-2022"> <meta name="generator" content="MediaWiki 1.44.0-wmf.4"> <meta name="referrer" content="origin"> <meta name="referrer" content="origin-when-cross-origin"> <meta name="robots" content="max-image-preview:standard"> <meta name="format-detection" content="telephone=no"> <meta name="viewport" content="width=1120"> <meta property="og:title" content="Convolutional neural network - Wikipedia"> <meta property="og:type" content="website"> <link rel="preconnect" href="//upload.wikimedia.org"> <link rel="alternate" media="only screen and (max-width: 640px)" href="//en.m.wikipedia.org/wiki/Convolutional_neural_network"> <link rel="alternate" type="application/x-wiki" title="Edit this page" href="/w/index.php?title=Convolutional_neural_network&action=edit"> <link rel="apple-touch-icon" href="/static/apple-touch/wikipedia.png"> <link rel="icon" href="/static/favicon/wikipedia.ico"> <link rel="search" type="application/opensearchdescription+xml" href="/w/rest.php/v1/search" title="Wikipedia (en)"> <link rel="EditURI" type="application/rsd+xml" href="//en.wikipedia.org/w/api.php?action=rsd"> <link rel="canonical" href="https://en.wikipedia.org/wiki/Convolutional_neural_network"> <link rel="license" href="https://creativecommons.org/licenses/by-sa/4.0/deed.en"> <link rel="alternate" type="application/atom+xml" title="Wikipedia Atom feed" href="/w/index.php?title=Special:RecentChanges&feed=atom"> <link rel="dns-prefetch" href="//meta.wikimedia.org" /> <link rel="dns-prefetch" href="//login.wikimedia.org"> </head> <body class="skin--responsive skin-vector skin-vector-search-vue mediawiki ltr sitedir-ltr mw-hide-empty-elt ns-0 ns-subject mw-editable page-Convolutional_neural_network rootpage-Convolutional_neural_network skin-vector-2022 action-view"><a class="mw-jump-link" href="#bodyContent">Jump to content</a> <div class="vector-header-container"> <header class="vector-header mw-header"> <div class="vector-header-start"> <nav class="vector-main-menu-landmark" aria-label="Site"> <div id="vector-main-menu-dropdown" class="vector-dropdown vector-main-menu-dropdown vector-button-flush-left vector-button-flush-right" > <input type="checkbox" id="vector-main-menu-dropdown-checkbox" role="button" aria-haspopup="true" data-event-name="ui.dropdown-vector-main-menu-dropdown" class="vector-dropdown-checkbox " aria-label="Main menu" > <label id="vector-main-menu-dropdown-label" for="vector-main-menu-dropdown-checkbox" class="vector-dropdown-label cdx-button cdx-button--fake-button cdx-button--fake-button--enabled cdx-button--weight-quiet cdx-button--icon-only " aria-hidden="true" > Main menu </label> <div class="vector-dropdown-content"> <div id="vector-main-menu-unpinned-container" class="vector-unpinned-container"> <div id="vector-main-menu" class="vector-main-menu vector-pinnable-element"> <div class="vector-pinnable-header vector-main-menu-pinnable-header vector-pinnable-header-unpinned" data-feature-name="main-menu-pinned" data-pinnable-element-id="vector-main-menu" data-pinned-container-id="vector-main-menu-pinned-container" data-unpinned-container-id="vector-main-menu-unpinned-container" > <div class="vector-pinnable-header-label">Main menu</div> <button class="vector-pinnable-header-toggle-button vector-pinnable-header-pin-button" data-event-name="pinnable-header.vector-main-menu.pin">move to sidebar</button> <button class="vector-pinnable-header-toggle-button vector-pinnable-header-unpin-button" data-event-name="pinnable-header.vector-main-menu.unpin">hide</button> </div> <div id="p-navigation" class="vector-menu mw-portlet mw-portlet-navigation" > <div class="vector-menu-heading"> Navigation </div> <div class="vector-menu-content"> <ul class="vector-menu-content-list"> <li id="n-mainpage-description" class="mw-list-item"><a href="/wiki/Main_Page" title="Visit the main page [z]" accesskey="z">Main page</a></li><li id="n-contents" class="mw-list-item"><a href="/wiki/Wikipedia:Contents" title="Guides to browsing Wikipedia">Contents</a></li><li id="n-currentevents" class="mw-list-item"><a href="/wiki/Portal:Current_events" title="Articles related to current events">Current events</a></li><li id="n-randompage" class="mw-list-item"><a href="/wiki/Special:Random" title="Visit a randomly selected article [x]" accesskey="x">Random article</a></li><li id="n-aboutsite" class="mw-list-item"><a href="/wiki/Wikipedia:About" title="Learn about Wikipedia and how it works">About Wikipedia</a></li><li id="n-contactpage" class="mw-list-item"><a href="//en.wikipedia.org/wiki/Wikipedia:Contact_us" title="How to contact Wikipedia">Contact us</a></li> </ul> </div> </div> <div id="p-interaction" class="vector-menu mw-portlet mw-portlet-interaction" > <div class="vector-menu-heading"> Contribute </div> <div class="vector-menu-content"> <ul class="vector-menu-content-list"> <li id="n-help" class="mw-list-item"><a href="/wiki/Help:Contents" title="Guidance on how to use and edit Wikipedia">Help</a></li><li id="n-introduction" class="mw-list-item"><a href="/wiki/Help:Introduction" title="Learn how to edit Wikipedia">Learn to edit</a></li><li id="n-portal" class="mw-list-item"><a href="/wiki/Wikipedia:Community_portal" title="The hub for editors">Community portal</a></li><li id="n-recentchanges" class="mw-list-item"><a href="/wiki/Special:RecentChanges" title="A list of recent changes to Wikipedia [r]" accesskey="r">Recent changes</a></li><li id="n-upload" class="mw-list-item"><a href="/wiki/Wikipedia:File_upload_wizard" title="Add images or other media for use on Wikipedia">Upload file</a></li> </ul> </div> </div> </div> </div> </div> </div> </nav> <a href="/wiki/Main_Page" class="mw-logo"> <img class="mw-logo-icon" src="/static/images/icons/wikipedia.png" alt="" aria-hidden="true" height="50" width="50"> <img class="mw-logo-wordmark" alt="Wikipedia" src="/static/images/mobile/copyright/wikipedia-wordmark-en.svg" style="width: 7.5em; height: 1.125em;"> <img class="mw-logo-tagline" alt="The Free Encyclopedia" src="/static/images/mobile/copyright/wikipedia-tagline-en.svg" width="117" height="13" style="width: 7.3125em; height: 0.8125em;"> </a> </div> <div class="vector-header-end"> <div id="p-search" role="search" class="vector-search-box-vue vector-search-box-collapses vector-search-box-show-thumbnail vector-search-box-auto-expand-width vector-search-box"> <a href="/wiki/Special:Search" class="cdx-button cdx-button--fake-button cdx-button--fake-button--enabled cdx-button--weight-quiet cdx-button--icon-only search-toggle" title="Search Wikipedia [f]" accesskey="f"> Search </a> <div class="vector-typeahead-search-container"> <div class="cdx-typeahead-search cdx-typeahead-search--show-thumbnail cdx-typeahead-search--auto-expand-width"> <form action="/w/index.php" id="searchform" class="cdx-search-input cdx-search-input--has-end-button"> <div id="simpleSearch" class="cdx-search-input__input-wrapper" data-search-loc="header-moved"> <div class="cdx-text-input cdx-text-input--has-start-icon"> <input class="cdx-text-input__input" type="search" name="search" placeholder="Search Wikipedia" aria-label="Search Wikipedia" autocapitalize="sentences" title="Search Wikipedia [f]" accesskey="f" id="searchInput" > </div> <input type="hidden" name="title" value="Special:Search"> </div> <button class="cdx-button cdx-search-input__end-button">Search</button> </form> </div> </div> </div> <nav class="vector-user-links vector-user-links-wide" aria-label="Personal tools"> <div class="vector-user-links-main"> <div id="p-vector-user-menu-preferences" class="vector-menu mw-portlet emptyPortlet" > <div class="vector-menu-content"> <ul class="vector-menu-content-list"> </ul> </div> </div> <div id="p-vector-user-menu-userpage" class="vector-menu mw-portlet emptyPortlet" > <div class="vector-menu-content"> <ul class="vector-menu-content-list"> </ul> </div> </div> <nav class="vector-appearance-landmark" aria-label="Appearance"> <div id="vector-appearance-dropdown" class="vector-dropdown " title="Change the appearance of the page's font size, width, and color" > <input type="checkbox" id="vector-appearance-dropdown-checkbox" role="button" aria-haspopup="true" data-event-name="ui.dropdown-vector-appearance-dropdown" class="vector-dropdown-checkbox " aria-label="Appearance" > <label id="vector-appearance-dropdown-label" for="vector-appearance-dropdown-checkbox" class="vector-dropdown-label cdx-button cdx-button--fake-button cdx-button--fake-button--enabled cdx-button--weight-quiet cdx-button--icon-only " aria-hidden="true" > Appearance </label> <div class="vector-dropdown-content"> <div id="vector-appearance-unpinned-container" class="vector-unpinned-container"> </div> </div> </div> </nav> <div id="p-vector-user-menu-notifications" class="vector-menu mw-portlet emptyPortlet" > <div class="vector-menu-content"> <ul class="vector-menu-content-list"> </ul> </div> </div> <div id="p-vector-user-menu-overflow" class="vector-menu mw-portlet" > <div class="vector-menu-content"> <ul class="vector-menu-content-list"> <li id="pt-sitesupport-2" class="user-links-collapsible-item mw-list-item user-links-collapsible-item"><a data-mw="interface" href="https://donate.wikimedia.org/wiki/Special:FundraiserRedirector?utm_source=donate&utm_medium=sidebar&utm_campaign=C13_en.wikipedia.org&uselang=en" class="">Donate</a> </li> <li id="pt-createaccount-2" class="user-links-collapsible-item mw-list-item user-links-collapsible-item"><a data-mw="interface" href="/w/index.php?title=Special:CreateAccount&returnto=Convolutional+neural+network" title="You are encouraged to create an account and log in; however, it is not mandatory" class="">Create account</a> </li> <li id="pt-login-2" class="user-links-collapsible-item mw-list-item user-links-collapsible-item"><a data-mw="interface" href="/w/index.php?title=Special:UserLogin&returnto=Convolutional+neural+network" title="You're encouraged to log in; however, it's not mandatory. [o]" accesskey="o" class="">Log in</a> </li> </ul> </div> </div> </div> <div id="vector-user-links-dropdown" class="vector-dropdown vector-user-menu vector-button-flush-right vector-user-menu-logged-out" title="Log in and more options" > <input type="checkbox" id="vector-user-links-dropdown-checkbox" role="button" aria-haspopup="true" data-event-name="ui.dropdown-vector-user-links-dropdown" class="vector-dropdown-checkbox " aria-label="Personal tools" > <label id="vector-user-links-dropdown-label" for="vector-user-links-dropdown-checkbox" class="vector-dropdown-label cdx-button cdx-button--fake-button cdx-button--fake-button--enabled cdx-button--weight-quiet cdx-button--icon-only " aria-hidden="true" > Personal tools </label> <div class="vector-dropdown-content"> <div id="p-personal" class="vector-menu mw-portlet mw-portlet-personal user-links-collapsible-item" title="User menu" > <div class="vector-menu-content"> <ul class="vector-menu-content-list"> <li id="pt-sitesupport" class="user-links-collapsible-item mw-list-item"><a href="https://donate.wikimedia.org/wiki/Special:FundraiserRedirector?utm_source=donate&utm_medium=sidebar&utm_campaign=C13_en.wikipedia.org&uselang=en">Donate</a></li><li id="pt-createaccount" class="user-links-collapsible-item mw-list-item"><a href="/w/index.php?title=Special:CreateAccount&returnto=Convolutional+neural+network" title="You are encouraged to create an account and log in; however, it is not mandatory"> Create account</a></li><li id="pt-login" class="user-links-collapsible-item mw-list-item"><a href="/w/index.php?title=Special:UserLogin&returnto=Convolutional+neural+network" title="You're encouraged to log in; however, it's not mandatory. [o]" accesskey="o"> Log in</a></li> </ul> </div> </div> <div id="p-user-menu-anon-editor" class="vector-menu mw-portlet mw-portlet-user-menu-anon-editor" > <div class="vector-menu-heading"> Pages for logged out editors <a href="/wiki/Help:Introduction" aria-label="Learn more about editing">learn more</a> </div> <div class="vector-menu-content"> <ul class="vector-menu-content-list"> <li id="pt-anoncontribs" class="mw-list-item"><a href="/wiki/Special:MyContributions" title="A list of edits made from this IP address [y]" accesskey="y">Contributions</a></li><li id="pt-anontalk" class="mw-list-item"><a href="/wiki/Special:MyTalk" title="Discussion about edits from this IP address [n]" accesskey="n">Talk</a></li> </ul> </div> </div> </div> </div> </nav> </div> </header> </div> <div class="mw-page-container"> <div class="mw-page-container-inner"> <div class="vector-sitenotice-container"> <div id="siteNotice"></div> </div> <div class="vector-column-start"> <div class="vector-main-menu-container"> <div id="mw-navigation"> <nav id="mw-panel" class="vector-main-menu-landmark" aria-label="Site"> <div id="vector-main-menu-pinned-container" class="vector-pinned-container"> </div> </nav> </div> </div> <div class="vector-sticky-pinned-container"> <nav id="mw-panel-toc" aria-label="Contents" data-event-name="ui.sidebar-toc" class="mw-table-of-contents-container vector-toc-landmark"> <div id="vector-toc-pinned-container" class="vector-pinned-container"> <div id="vector-toc" class="vector-toc vector-pinnable-element"> <div class="vector-pinnable-header vector-toc-pinnable-header vector-pinnable-header-pinned" data-feature-name="toc-pinned" data-pinnable-element-id="vector-toc" > <h2 class="vector-pinnable-header-label">Contents</h2> <button class="vector-pinnable-header-toggle-button vector-pinnable-header-pin-button" data-event-name="pinnable-header.vector-toc.pin">move to sidebar</button> <button class="vector-pinnable-header-toggle-button vector-pinnable-header-unpin-button" data-event-name="pinnable-header.vector-toc.unpin">hide</button> </div> <ul class="vector-toc-contents" id="mw-panel-toc-list"> <li id="toc-mw-content-text" class="vector-toc-list-item vector-toc-level-1"> <a href="#" class="vector-toc-link"> <div class="vector-toc-text">(Top)</div> </a> </li> <li id="toc-Architecture" class="vector-toc-list-item vector-toc-level-1"> <a class="vector-toc-link" href="#Architecture"> <div class="vector-toc-text"> 1 Architecture </div> </a> <button aria-controls="toc-Architecture-sublist" class="cdx-button cdx-button--weight-quiet cdx-button--icon-only vector-toc-toggle"> Toggle Architecture subsection </button> <ul id="toc-Architecture-sublist" class="vector-toc-list"> <li id="toc-Convolutional_layers" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Convolutional_layers"> <div class="vector-toc-text"> 1.1 Convolutional layers </div> </a> <ul id="toc-Convolutional_layers-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Pooling_layers" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Pooling_layers"> <div class="vector-toc-text"> 1.2 Pooling layers </div> </a> <ul id="toc-Pooling_layers-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Fully_connected_layers" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Fully_connected_layers"> <div class="vector-toc-text"> 1.3 Fully connected layers </div> </a> <ul id="toc-Fully_connected_layers-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Receptive_field" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Receptive_field"> <div class="vector-toc-text"> 1.4 Receptive field </div> </a> <ul id="toc-Receptive_field-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Weights" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Weights"> <div class="vector-toc-text"> 1.5 Weights </div> </a> <ul id="toc-Weights-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Deconvolutional" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Deconvolutional"> <div class="vector-toc-text"> 1.6 Deconvolutional </div> </a> <ul id="toc-Deconvolutional-sublist" class="vector-toc-list"> </ul> </li> </ul> </li> <li id="toc-History" class="vector-toc-list-item vector-toc-level-1"> <a class="vector-toc-link" href="#History"> <div class="vector-toc-text"> 2 History </div> </a> <button aria-controls="toc-History-sublist" class="cdx-button cdx-button--weight-quiet cdx-button--icon-only vector-toc-toggle"> Toggle History subsection </button> <ul id="toc-History-sublist" class="vector-toc-list"> <li id="toc-Receptive_fields_in_the_visual_cortex" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Receptive_fields_in_the_visual_cortex"> <div class="vector-toc-text"> 2.1 Receptive fields in the visual cortex </div> </a> <ul id="toc-Receptive_fields_in_the_visual_cortex-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Neocognitron,_origin_of_the_CNN_architecture" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Neocognitron,_origin_of_the_CNN_architecture"> <div class="vector-toc-text"> 2.2 Neocognitron, origin of the CNN architecture </div> </a> <ul id="toc-Neocognitron,_origin_of_the_CNN_architecture-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Convolution_in_time" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Convolution_in_time"> <div class="vector-toc-text"> 2.3 Convolution in time </div> </a> <ul id="toc-Convolution_in_time-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Time_delay_neural_networks" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Time_delay_neural_networks"> <div class="vector-toc-text"> 2.4 Time delay neural networks </div> </a> <ul id="toc-Time_delay_neural_networks-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Image_recognition_with_CNNs_trained_by_gradient_descent" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Image_recognition_with_CNNs_trained_by_gradient_descent"> <div class="vector-toc-text"> 2.5 Image recognition with CNNs trained by gradient descent </div> </a> <ul id="toc-Image_recognition_with_CNNs_trained_by_gradient_descent-sublist" class="vector-toc-list"> <li id="toc-Max_pooling" class="vector-toc-list-item vector-toc-level-3"> <a class="vector-toc-link" href="#Max_pooling"> <div class="vector-toc-text"> 2.5.1 Max pooling </div> </a> <ul id="toc-Max_pooling-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-LeNet-5" class="vector-toc-list-item vector-toc-level-3"> <a class="vector-toc-link" href="#LeNet-5"> <div class="vector-toc-text"> 2.5.2 LeNet-5 </div> </a> <ul id="toc-LeNet-5-sublist" class="vector-toc-list"> </ul> </li> </ul> </li> <li id="toc-Shift-invariant_neural_network" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Shift-invariant_neural_network"> <div class="vector-toc-text"> 2.6 Shift-invariant neural network </div> </a> <ul id="toc-Shift-invariant_neural_network-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-GPU_implementations" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#GPU_implementations"> <div class="vector-toc-text"> 2.7 GPU implementations </div> </a> <ul id="toc-GPU_implementations-sublist" class="vector-toc-list"> </ul> </li> </ul> </li> <li id="toc-Distinguishing_features" class="vector-toc-list-item vector-toc-level-1"> <a class="vector-toc-link" href="#Distinguishing_features"> <div class="vector-toc-text"> 3 Distinguishing features </div> </a> <ul id="toc-Distinguishing_features-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Building_blocks" class="vector-toc-list-item vector-toc-level-1"> <a class="vector-toc-link" href="#Building_blocks"> <div class="vector-toc-text"> 4 Building blocks </div> </a> <button aria-controls="toc-Building_blocks-sublist" class="cdx-button cdx-button--weight-quiet cdx-button--icon-only vector-toc-toggle"> Toggle Building blocks subsection </button> <ul id="toc-Building_blocks-sublist" class="vector-toc-list"> <li id="toc-Convolutional_layer" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Convolutional_layer"> <div class="vector-toc-text"> 4.1 Convolutional layer </div> </a> <ul id="toc-Convolutional_layer-sublist" class="vector-toc-list"> <li id="toc-Local_connectivity" class="vector-toc-list-item vector-toc-level-3"> <a class="vector-toc-link" href="#Local_connectivity"> <div class="vector-toc-text"> 4.1.1 Local connectivity </div> </a> <ul id="toc-Local_connectivity-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Spatial_arrangement" class="vector-toc-list-item vector-toc-level-3"> <a class="vector-toc-link" href="#Spatial_arrangement"> <div class="vector-toc-text"> 4.1.2 Spatial arrangement </div> </a> <ul id="toc-Spatial_arrangement-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Parameter_sharing" class="vector-toc-list-item vector-toc-level-3"> <a class="vector-toc-link" href="#Parameter_sharing"> <div class="vector-toc-text"> 4.1.3 Parameter sharing </div> </a> <ul id="toc-Parameter_sharing-sublist" class="vector-toc-list"> </ul> </li> </ul> </li> <li id="toc-Pooling_layer" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Pooling_layer"> <div class="vector-toc-text"> 4.2 Pooling layer </div> </a> <ul id="toc-Pooling_layer-sublist" class="vector-toc-list"> <li id="toc-Channel_max_pooling" class="vector-toc-list-item vector-toc-level-3"> <a class="vector-toc-link" href="#Channel_max_pooling"> <div class="vector-toc-text"> 4.2.1 Channel max pooling </div> </a> <ul id="toc-Channel_max_pooling-sublist" class="vector-toc-list"> </ul> </li> </ul> </li> <li id="toc-ReLU_layer" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#ReLU_layer"> <div class="vector-toc-text"> 4.3 ReLU layer </div> </a> <ul id="toc-ReLU_layer-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Fully_connected_layer" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Fully_connected_layer"> <div class="vector-toc-text"> 4.4 Fully connected layer </div> </a> <ul id="toc-Fully_connected_layer-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Loss_layer" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Loss_layer"> <div class="vector-toc-text"> 4.5 Loss layer </div> </a> <ul id="toc-Loss_layer-sublist" class="vector-toc-list"> </ul> </li> </ul> </li> <li id="toc-Hyperparameters" class="vector-toc-list-item vector-toc-level-1"> <a class="vector-toc-link" href="#Hyperparameters"> <div class="vector-toc-text"> 5 Hyperparameters </div> </a> <button aria-controls="toc-Hyperparameters-sublist" class="cdx-button cdx-button--weight-quiet cdx-button--icon-only vector-toc-toggle"> Toggle Hyperparameters subsection </button> <ul id="toc-Hyperparameters-sublist" class="vector-toc-list"> <li id="toc-Kernel_size" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Kernel_size"> <div class="vector-toc-text"> 5.1 Kernel size </div> </a> <ul id="toc-Kernel_size-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Padding" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Padding"> <div class="vector-toc-text"> 5.2 Padding </div> </a> <ul id="toc-Padding-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Stride" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Stride"> <div class="vector-toc-text"> 5.3 Stride </div> </a> <ul id="toc-Stride-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Number_of_filters" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Number_of_filters"> <div class="vector-toc-text"> 5.4 Number of filters </div> </a> <ul id="toc-Number_of_filters-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Filter_size" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Filter_size"> <div class="vector-toc-text"> 5.5 Filter size </div> </a> <ul id="toc-Filter_size-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Pooling_type_and_size" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Pooling_type_and_size"> <div class="vector-toc-text"> 5.6 Pooling type and size </div> </a> <ul id="toc-Pooling_type_and_size-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Dilation" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Dilation"> <div class="vector-toc-text"> 5.7 Dilation </div> </a> <ul id="toc-Dilation-sublist" class="vector-toc-list"> </ul> </li> </ul> </li> <li id="toc-Translation_equivariance_and_aliasing" class="vector-toc-list-item vector-toc-level-1"> <a class="vector-toc-link" href="#Translation_equivariance_and_aliasing"> <div class="vector-toc-text"> 6 Translation equivariance and aliasing </div> </a> <ul id="toc-Translation_equivariance_and_aliasing-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Evaluation" class="vector-toc-list-item vector-toc-level-1"> <a class="vector-toc-link" href="#Evaluation"> <div class="vector-toc-text"> 7 Evaluation </div> </a> <ul id="toc-Evaluation-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Regularization_methods" class="vector-toc-list-item vector-toc-level-1"> <a class="vector-toc-link" href="#Regularization_methods"> <div class="vector-toc-text"> 8 Regularization methods </div> </a> <button aria-controls="toc-Regularization_methods-sublist" class="cdx-button cdx-button--weight-quiet cdx-button--icon-only vector-toc-toggle"> Toggle Regularization methods subsection </button> <ul id="toc-Regularization_methods-sublist" class="vector-toc-list"> <li id="toc-Empirical" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Empirical"> <div class="vector-toc-text"> 8.1 Empirical </div> </a> <ul id="toc-Empirical-sublist" class="vector-toc-list"> <li id="toc-Dropout" class="vector-toc-list-item vector-toc-level-3"> <a class="vector-toc-link" href="#Dropout"> <div class="vector-toc-text"> 8.1.1 Dropout </div> </a> <ul id="toc-Dropout-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-DropConnect" class="vector-toc-list-item vector-toc-level-3"> <a class="vector-toc-link" href="#DropConnect"> <div class="vector-toc-text"> 8.1.2 DropConnect </div> </a> <ul id="toc-DropConnect-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Stochastic_pooling" class="vector-toc-list-item vector-toc-level-3"> <a class="vector-toc-link" href="#Stochastic_pooling"> <div class="vector-toc-text"> 8.1.3 Stochastic pooling </div> </a> <ul id="toc-Stochastic_pooling-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Artificial_data" class="vector-toc-list-item vector-toc-level-3"> <a class="vector-toc-link" href="#Artificial_data"> <div class="vector-toc-text"> 8.1.4 Artificial data </div> </a> <ul id="toc-Artificial_data-sublist" class="vector-toc-list"> </ul> </li> </ul> </li> <li id="toc-Explicit" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Explicit"> <div class="vector-toc-text"> 8.2 Explicit </div> </a> <ul id="toc-Explicit-sublist" class="vector-toc-list"> <li id="toc-Early_stopping" class="vector-toc-list-item vector-toc-level-3"> <a class="vector-toc-link" href="#Early_stopping"> <div class="vector-toc-text"> 8.2.1 Early stopping </div> </a> <ul id="toc-Early_stopping-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Number_of_parameters" class="vector-toc-list-item vector-toc-level-3"> <a class="vector-toc-link" href="#Number_of_parameters"> <div class="vector-toc-text"> 8.2.2 Number of parameters </div> </a> <ul id="toc-Number_of_parameters-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Weight_decay" class="vector-toc-list-item vector-toc-level-3"> <a class="vector-toc-link" href="#Weight_decay"> <div class="vector-toc-text"> 8.2.3 Weight decay </div> </a> <ul id="toc-Weight_decay-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Max_norm_constraints" class="vector-toc-list-item vector-toc-level-3"> <a class="vector-toc-link" href="#Max_norm_constraints"> <div class="vector-toc-text"> 8.2.4 Max norm constraints </div> </a> <ul id="toc-Max_norm_constraints-sublist" class="vector-toc-list"> </ul> </li> </ul> </li> </ul> </li> <li id="toc-Hierarchical_coordinate_frames" class="vector-toc-list-item vector-toc-level-1"> <a class="vector-toc-link" href="#Hierarchical_coordinate_frames"> <div class="vector-toc-text"> 9 Hierarchical coordinate frames </div> </a> <ul id="toc-Hierarchical_coordinate_frames-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Applications" class="vector-toc-list-item vector-toc-level-1"> <a class="vector-toc-link" href="#Applications"> <div class="vector-toc-text"> 10 Applications </div> </a> <button aria-controls="toc-Applications-sublist" class="cdx-button cdx-button--weight-quiet cdx-button--icon-only vector-toc-toggle"> Toggle Applications subsection </button> <ul id="toc-Applications-sublist" class="vector-toc-list"> <li id="toc-Image_recognition" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Image_recognition"> <div class="vector-toc-text"> 10.1 Image recognition </div> </a> <ul id="toc-Image_recognition-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Video_analysis" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Video_analysis"> <div class="vector-toc-text"> 10.2 Video analysis </div> </a> <ul id="toc-Video_analysis-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Natural_language_processing" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Natural_language_processing"> <div class="vector-toc-text"> 10.3 Natural language processing </div> </a> <ul id="toc-Natural_language_processing-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Anomaly_detection" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Anomaly_detection"> <div class="vector-toc-text"> 10.4 Anomaly detection </div> </a> <ul id="toc-Anomaly_detection-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Drug_discovery" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Drug_discovery"> <div class="vector-toc-text"> 10.5 Drug discovery </div> </a> <ul id="toc-Drug_discovery-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Checkers_game" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Checkers_game"> <div class="vector-toc-text"> 10.6 Checkers game </div> </a> <ul id="toc-Checkers_game-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Go" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Go"> <div class="vector-toc-text"> 10.7 Go </div> </a> <ul id="toc-Go-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Time_series_forecasting" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Time_series_forecasting"> <div class="vector-toc-text"> 10.8 Time series forecasting </div> </a> <ul id="toc-Time_series_forecasting-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Cultural_heritage_and_3D-datasets" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Cultural_heritage_and_3D-datasets"> <div class="vector-toc-text"> 10.9 Cultural heritage and 3D-datasets </div> </a> <ul id="toc-Cultural_heritage_and_3D-datasets-sublist" class="vector-toc-list"> </ul> </li> </ul> </li> <li id="toc-Fine-tuning" class="vector-toc-list-item vector-toc-level-1"> <a class="vector-toc-link" href="#Fine-tuning"> <div class="vector-toc-text"> 11 Fine-tuning </div> </a> <ul id="toc-Fine-tuning-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Human_interpretable_explanations" class="vector-toc-list-item vector-toc-level-1"> <a class="vector-toc-link" href="#Human_interpretable_explanations"> <div class="vector-toc-text"> 12 Human interpretable explanations </div> </a> <ul id="toc-Human_interpretable_explanations-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Related_architectures" class="vector-toc-list-item vector-toc-level-1"> <a class="vector-toc-link" href="#Related_architectures"> <div class="vector-toc-text"> 13 Related architectures </div> </a> <button aria-controls="toc-Related_architectures-sublist" class="cdx-button cdx-button--weight-quiet cdx-button--icon-only vector-toc-toggle"> Toggle Related architectures subsection </button> <ul id="toc-Related_architectures-sublist" class="vector-toc-list"> <li id="toc-Deep_Q-networks" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Deep_Q-networks"> <div class="vector-toc-text"> 13.1 Deep Q-networks </div> </a> <ul id="toc-Deep_Q-networks-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Deep_belief_networks" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Deep_belief_networks"> <div class="vector-toc-text"> 13.2 Deep belief networks </div> </a> <ul id="toc-Deep_belief_networks-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Neural_abstraction_pyramid" class="vector-toc-list-item vector-toc-level-2"> <a class="vector-toc-link" href="#Neural_abstraction_pyramid"> <div class="vector-toc-text"> 13.3 Neural abstraction pyramid </div> </a> <ul id="toc-Neural_abstraction_pyramid-sublist" class="vector-toc-list"> </ul> </li> </ul> </li> <li id="toc-Notable_libraries" class="vector-toc-list-item vector-toc-level-1"> <a class="vector-toc-link" href="#Notable_libraries"> <div class="vector-toc-text"> 14 Notable libraries </div> </a> <ul id="toc-Notable_libraries-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-See_also" class="vector-toc-list-item vector-toc-level-1"> <a class="vector-toc-link" href="#See_also"> <div class="vector-toc-text"> 15 See also </div> </a> <ul id="toc-See_also-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-Notes" class="vector-toc-list-item vector-toc-level-1"> <a class="vector-toc-link" href="#Notes"> <div class="vector-toc-text"> 16 Notes </div> </a> <ul id="toc-Notes-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-References" class="vector-toc-list-item vector-toc-level-1"> <a class="vector-toc-link" href="#References"> <div class="vector-toc-text"> 17 References </div> </a> <ul id="toc-References-sublist" class="vector-toc-list"> </ul> </li> <li id="toc-External_links" class="vector-toc-list-item vector-toc-level-1"> <a class="vector-toc-link" href="#External_links"> <div class="vector-toc-text"> 18 External links </div> </a> <ul id="toc-External_links-sublist" class="vector-toc-list"> </ul> </li> </ul> </div> </div> </nav> </div> </div> <div class="mw-content-container"> <main id="content" class="mw-body"> <header class="mw-body-header vector-page-titlebar"> <nav aria-label="Contents" class="vector-toc-landmark"> <div id="vector-page-titlebar-toc" class="vector-dropdown vector-page-titlebar-toc vector-button-flush-left" > <input type="checkbox" id="vector-page-titlebar-toc-checkbox" role="button" aria-haspopup="true" data-event-name="ui.dropdown-vector-page-titlebar-toc" class="vector-dropdown-checkbox " aria-label="Toggle the table of contents" > <label id="vector-page-titlebar-toc-label" for="vector-page-titlebar-toc-checkbox" class="vector-dropdown-label cdx-button cdx-button--fake-button cdx-button--fake-button--enabled cdx-button--weight-quiet cdx-button--icon-only " aria-hidden="true" > Toggle the table of contents </label> <div class="vector-dropdown-content"> <div id="vector-page-titlebar-toc-unpinned-container" class="vector-unpinned-container"> </div> </div> </div> </nav> <h1 id="firstHeading" class="firstHeading mw-first-heading">Convolutional neural network</h1> <div id="p-lang-btn" class="vector-dropdown mw-portlet mw-portlet-lang" > <input type="checkbox" id="p-lang-btn-checkbox" role="button" aria-haspopup="true" data-event-name="ui.dropdown-p-lang-btn" class="vector-dropdown-checkbox mw-interlanguage-selector" aria-label="Go to an article in another language. Available in 28 languages" > <label id="p-lang-btn-label" for="p-lang-btn-checkbox" class="vector-dropdown-label cdx-button cdx-button--fake-button cdx-button--fake-button--enabled cdx-button--weight-quiet cdx-button--action-progressive mw-portlet-lang-heading-28" aria-hidden="true" > 28 languages </label> <div class="vector-dropdown-content"> <div class="vector-menu-content"> <ul class="vector-menu-content-list"> <li class="interlanguage-link interwiki-ar mw-list-item"><a href="https://ar.wikipedia.org/wiki/%D8%B4%D8%A8%D9%83%D8%A9_%D8%B9%D8%B5%D8%A8%D9%88%D9%86%D9%8A%D8%A9_%D8%A7%D9%84%D8%AA%D9%81%D8%A7%D9%81%D9%8A%D8%A9" title="شبكة عصبونية التفافية – Arabic" lang="ar" hreflang="ar" data-title="شبكة عصبونية التفافية" data-language-autonym="العربية" data-language-local-name="Arabic" class="interlanguage-link-target">العربية</a></li><li class="interlanguage-link interwiki-azb mw-list-item"><a href="https://azb.wikipedia.org/wiki/%D8%A7%D8%A6%D9%88%D8%B1%DB%8C%D8%B4%DB%8C%D9%85%D9%84%DB%8C_%D8%B9%D8%B5%D8%A8%DB%8C_%D8%B4%D8%A8%DA%A9%D9%87%E2%80%8C%D8%B3%DB%8C" title="ائوریشیملی عصبی شبکه‌سی – South Azerbaijani" lang="azb" hreflang="azb" data-title="ائوریشیملی عصبی شبکه‌سی" data-language-autonym="تۆرکجه" data-language-local-name="South Azerbaijani" class="interlanguage-link-target">تۆرکجه</a></li><li class="interlanguage-link interwiki-ca mw-list-item"><a href="https://ca.wikipedia.org/wiki/Xarxa_neuronal_convolutiva" title="Xarxa neuronal convolutiva – Catalan" lang="ca" hreflang="ca" data-title="Xarxa neuronal convolutiva" data-language-autonym="Català" data-language-local-name="Catalan" class="interlanguage-link-target">Català</a></li><li class="interlanguage-link interwiki-de mw-list-item"><a href="https://de.wikipedia.org/wiki/Convolutional_Neural_Network" title="Convolutional Neural Network – German" lang="de" hreflang="de" data-title="Convolutional Neural Network" data-language-autonym="Deutsch" data-language-local-name="German" class="interlanguage-link-target">Deutsch</a></li><li class="interlanguage-link interwiki-et mw-list-item"><a href="https://et.wikipedia.org/wiki/Konvolutsiooniline_n%C3%A4rviv%C3%B5rk" title="Konvolutsiooniline närvivõrk – Estonian" lang="et" hreflang="et" data-title="Konvolutsiooniline närvivõrk" data-language-autonym="Eesti" data-language-local-name="Estonian" class="interlanguage-link-target">Eesti</a></li><li class="interlanguage-link interwiki-es mw-list-item"><a href="https://es.wikipedia.org/wiki/Red_neuronal_convolucional" title="Red neuronal convolucional – Spanish" lang="es" hreflang="es" data-title="Red neuronal convolucional" data-language-autonym="Español" data-language-local-name="Spanish" class="interlanguage-link-target">Español</a></li><li class="interlanguage-link interwiki-eu mw-list-item"><a href="https://eu.wikipedia.org/wiki/Neurona-sare_konboluzional" title="Neurona-sare konboluzional – Basque" lang="eu" hreflang="eu" data-title="Neurona-sare konboluzional" data-language-autonym="Euskara" data-language-local-name="Basque" class="interlanguage-link-target">Euskara</a></li><li class="interlanguage-link interwiki-fa mw-list-item"><a href="https://fa.wikipedia.org/wiki/%D8%B4%D8%A8%DA%A9%D9%87_%D8%B9%D8%B5%D8%A8%DB%8C_%D9%BE%DB%8C%DA%86%D8%B4%DB%8C" title="شبکه عصبی پیچشی – Persian" lang="fa" hreflang="fa" data-title="شبکه عصبی پیچشی" data-language-autonym="فارسی" data-language-local-name="Persian" class="interlanguage-link-target">فارسی</a></li><li class="interlanguage-link interwiki-fr mw-list-item"><a href="https://fr.wikipedia.org/wiki/R%C3%A9seau_neuronal_convolutif" title="Réseau neuronal convolutif – French" lang="fr" hreflang="fr" data-title="Réseau neuronal convolutif" data-language-autonym="Français" data-language-local-name="French" class="interlanguage-link-target">Français</a></li><li class="interlanguage-link interwiki-gl mw-list-item"><a href="https://gl.wikipedia.org/wiki/Rede_neural_convolucional" title="Rede neural convolucional – Galician" lang="gl" hreflang="gl" data-title="Rede neural convolucional" data-language-autonym="Galego" data-language-local-name="Galician" class="interlanguage-link-target">Galego</a></li><li class="interlanguage-link interwiki-ko mw-list-item"><a href="https://ko.wikipedia.org/wiki/%ED%95%A9%EC%84%B1%EA%B3%B1_%EC%8B%A0%EA%B2%BD%EB%A7%9D" title="합성곱 신경망 – Korean" lang="ko" hreflang="ko" data-title="합성곱 신경망" data-language-autonym="한국어" data-language-local-name="Korean" class="interlanguage-link-target">한국어</a></li><li class="interlanguage-link interwiki-id mw-list-item"><a href="https://id.wikipedia.org/wiki/Jaringan_saraf_konvolusional" title="Jaringan saraf konvolusional – Indonesian" lang="id" hreflang="id" data-title="Jaringan saraf konvolusional" data-language-autonym="Bahasa Indonesia" data-language-local-name="Indonesian" class="interlanguage-link-target">Bahasa Indonesia</a></li><li class="interlanguage-link interwiki-it mw-list-item"><a href="https://it.wikipedia.org/wiki/Rete_neurale_convoluzionale" title="Rete neurale convoluzionale – Italian" lang="it" hreflang="it" data-title="Rete neurale convoluzionale" data-language-autonym="Italiano" data-language-local-name="Italian" class="interlanguage-link-target">Italiano</a></li><li class="interlanguage-link interwiki-he mw-list-item"><a href="https://he.wikipedia.org/wiki/%D7%A8%D7%A9%D7%AA_%D7%A7%D7%95%D7%A0%D7%91%D7%95%D7%9C%D7%95%D7%A6%D7%99%D7%94" title="רשת קונבולוציה – Hebrew" lang="he" hreflang="he" data-title="רשת קונבולוציה" data-language-autonym="עברית" data-language-local-name="Hebrew" class="interlanguage-link-target">עברית</a></li><li class="interlanguage-link interwiki-lt mw-list-item"><a href="https://lt.wikipedia.org/wiki/Konvoliucinis_neuroninis_tinklas" title="Konvoliucinis neuroninis tinklas – Lithuanian" lang="lt" hreflang="lt" data-title="Konvoliucinis neuroninis tinklas" data-language-autonym="Lietuvių" data-language-local-name="Lithuanian" class="interlanguage-link-target">Lietuvių</a></li><li class="interlanguage-link interwiki-ja mw-list-item"><a href="https://ja.wikipedia.org/wiki/%E7%95%B3%E3%81%BF%E8%BE%BC%E3%81%BF%E3%83%8B%E3%83%A5%E3%83%BC%E3%83%A9%E3%83%AB%E3%83%8D%E3%83%83%E3%83%88%E3%83%AF%E3%83%BC%E3%82%AF" title="畳み込みニューラルネットワーク – Japanese" lang="ja" hreflang="ja" data-title="畳み込みニューラルネットワーク" data-language-autonym="日本語" data-language-local-name="Japanese" class="interlanguage-link-target">日本語</a></li><li class="interlanguage-link interwiki-pt mw-list-item"><a href="https://pt.wikipedia.org/wiki/Rede_neural_convolucional" title="Rede neural convolucional – Portuguese" lang="pt" hreflang="pt" data-title="Rede neural convolucional" data-language-autonym="Português" data-language-local-name="Portuguese" class="interlanguage-link-target">Português</a></li><li class="interlanguage-link interwiki-qu mw-list-item"><a href="https://qu.wikipedia.org/wiki/K%27uyukuq_ankucha_llika" title="K'uyukuq ankucha llika – Quechua" lang="qu" hreflang="qu" data-title="K'uyukuq ankucha llika" data-language-autonym="Runa Simi" data-language-local-name="Quechua" class="interlanguage-link-target">Runa Simi</a></li><li class="interlanguage-link interwiki-ru mw-list-item"><a href="https://ru.wikipedia.org/wiki/%D0%A1%D0%B2%D1%91%D1%80%D1%82%D0%BE%D1%87%D0%BD%D0%B0%D1%8F_%D0%BD%D0%B5%D0%B9%D1%80%D0%BE%D0%BD%D0%BD%D0%B0%D1%8F_%D1%81%D0%B5%D1%82%D1%8C" title="Свёрточная нейронная сеть – Russian" lang="ru" hreflang="ru" data-title="Свёрточная нейронная сеть" data-language-autonym="Русский" data-language-local-name="Russian" class="interlanguage-link-target">Русский</a></li><li class="interlanguage-link interwiki-simple mw-list-item"><a href="https://simple.wikipedia.org/wiki/Convolutional_neural_network" title="Convolutional neural network – Simple English" lang="en-simple" hreflang="en-simple" data-title="Convolutional neural network" data-language-autonym="Simple English" data-language-local-name="Simple English" class="interlanguage-link-target">Simple English</a></li><li class="interlanguage-link interwiki-sr mw-list-item"><a href="https://sr.wikipedia.org/wiki/%D0%9A%D0%BE%D0%BD%D0%B2%D0%BE%D0%BB%D1%83%D1%86%D0%B8%D1%98%D1%81%D0%BA%D0%B5_%D0%BD%D0%B5%D1%83%D1%80%D0%BE%D0%BD%D1%81%D0%BA%D0%B5_%D0%BC%D1%80%D0%B5%D0%B6%D0%B5" title="Конволуцијске неуронске мреже – Serbian" lang="sr" hreflang="sr" data-title="Конволуцијске неуронске мреже" data-language-autonym="Српски / srpski" data-language-local-name="Serbian" class="interlanguage-link-target">Српски / srpski</a></li><li class="interlanguage-link interwiki-th mw-list-item"><a href="https://th.wikipedia.org/wiki/%E0%B9%82%E0%B8%84%E0%B8%A3%E0%B8%87%E0%B8%82%E0%B9%88%E0%B8%B2%E0%B8%A2%E0%B8%9B%E0%B8%A3%E0%B8%B0%E0%B8%AA%E0%B8%B2%E0%B8%97%E0%B9%81%E0%B8%9A%E0%B8%9A%E0%B8%AA%E0%B8%B1%E0%B8%87%E0%B8%A7%E0%B8%B1%E0%B8%95%E0%B8%99%E0%B8%B2%E0%B8%81%E0%B8%B2%E0%B8%A3" title="โครงข่ายประสาทแบบสังวัตนาการ – Thai" lang="th" hreflang="th" data-title="โครงข่ายประสาทแบบสังวัตนาการ" data-language-autonym="ไทย" data-language-local-name="Thai" class="interlanguage-link-target">ไทย</a></li><li class="interlanguage-link interwiki-tr mw-list-item"><a href="https://tr.wikipedia.org/wiki/Evri%C5%9Fimli_sinir_a%C4%9Flar%C4%B1" title="Evrişimli sinir ağları – Turkish" lang="tr" hreflang="tr" data-title="Evrişimli sinir ağları" data-language-autonym="Türkçe" data-language-local-name="Turkish" class="interlanguage-link-target">Türkçe</a></li><li class="interlanguage-link interwiki-uk mw-list-item"><a href="https://uk.wikipedia.org/wiki/%D0%97%D0%B3%D0%BE%D1%80%D1%82%D0%BA%D0%BE%D0%B2%D0%B0_%D0%BD%D0%B5%D0%B9%D1%80%D0%BE%D0%BD%D0%BD%D0%B0_%D0%BC%D0%B5%D1%80%D0%B5%D0%B6%D0%B0" title="Згорткова нейронна мережа – Ukrainian" lang="uk" hreflang="uk" data-title="Згорткова нейронна мережа" data-language-autonym="Українська" data-language-local-name="Ukrainian" class="interlanguage-link-target">Українська</a></li><li class="interlanguage-link interwiki-vi mw-list-item"><a href="https://vi.wikipedia.org/wiki/M%E1%BA%A1ng_th%E1%BA%A7n_kinh_t%C3%ADch_ch%E1%BA%ADp" title="Mạng thần kinh tích chập – Vietnamese" lang="vi" hreflang="vi" data-title="Mạng thần kinh tích chập" data-language-autonym="Tiếng Việt" data-language-local-name="Vietnamese" class="interlanguage-link-target">Tiếng Việt</a></li><li class="interlanguage-link interwiki-wuu mw-list-item"><a href="https://wuu.wikipedia.org/wiki/%E5%8D%B7%E7%A7%AF%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C" title="卷积神经网络 – Wu" lang="wuu" hreflang="wuu" data-title="卷积神经网络" data-language-autonym="吴语" data-language-local-name="Wu" class="interlanguage-link-target">吴语</a></li><li class="interlanguage-link interwiki-zh-yue mw-list-item"><a href="https://zh-yue.wikipedia.org/wiki/%E5%8D%B7%E7%A9%8D%E7%A5%9E%E7%B6%93%E7%B6%B2%E7%B5%A1" title="卷積神經網絡 – Cantonese" lang="yue" hreflang="yue" data-title="卷積神經網絡" data-language-autonym="粵語" data-language-local-name="Cantonese" class="interlanguage-link-target">粵語</a></li><li class="interlanguage-link interwiki-zh mw-list-item"><a href="https://zh.wikipedia.org/wiki/%E5%8D%B7%E7%A7%AF%E7%A5%9E%E7%BB%8F%E7%BD%91%E7%BB%9C" title="卷积神经网络 – Chinese" lang="zh" hreflang="zh" data-title="卷积神经网络" data-language-autonym="中文" data-language-local-name="Chinese" class="interlanguage-link-target">中文</a></li> </ul> <div class="after-portlet after-portlet-lang"><a href="https://www.wikidata.org/wiki/Special:EntityPage/Q17084460#sitelinks-wikipedia" title="Edit interlanguage links" class="wbc-editpage">Edit links</a></div> </div> </div> </div> </header> <div class="vector-page-toolbar"> <div class="vector-page-toolbar-container"> <div id="left-navigation"> <nav aria-label="Namespaces"> <div id="p-associated-pages" class="vector-menu vector-menu-tabs mw-portlet mw-portlet-associated-pages" > <div class="vector-menu-content"> <ul class="vector-menu-content-list"> <li id="ca-nstab-main" class="selected vector-tab-noicon mw-list-item"><a href="/wiki/Convolutional_neural_network" title="View the content page [c]" accesskey="c">Article</a></li><li id="ca-talk" class="vector-tab-noicon mw-list-item"><a href="/wiki/Talk:Convolutional_neural_network" rel="discussion" title="Discuss improvements to the content page [t]" accesskey="t">Talk</a></li> </ul> </div> </div> <div id="vector-variants-dropdown" class="vector-dropdown emptyPortlet" > <input type="checkbox" id="vector-variants-dropdown-checkbox" role="button" aria-haspopup="true" data-event-name="ui.dropdown-vector-variants-dropdown" class="vector-dropdown-checkbox " aria-label="Change language variant" > <label id="vector-variants-dropdown-label" for="vector-variants-dropdown-checkbox" class="vector-dropdown-label cdx-button cdx-button--fake-button cdx-button--fake-button--enabled cdx-button--weight-quiet" aria-hidden="true" >English </label> <div class="vector-dropdown-content"> <div id="p-variants" class="vector-menu mw-portlet mw-portlet-variants emptyPortlet" > <div class="vector-menu-content"> <ul class="vector-menu-content-list"> </ul> </div> </div> </div> </div> </nav> </div> <div id="right-navigation" class="vector-collapsible"> <nav aria-label="Views"> <div id="p-views" class="vector-menu vector-menu-tabs mw-portlet mw-portlet-views" > <div class="vector-menu-content"> <ul class="vector-menu-content-list"> <li id="ca-view" class="selected vector-tab-noicon mw-list-item"><a href="/wiki/Convolutional_neural_network">Read</a></li><li id="ca-edit" class="vector-tab-noicon mw-list-item"><a href="/w/index.php?title=Convolutional_neural_network&action=edit" title="Edit this page [e]" accesskey="e">Edit</a></li><li id="ca-history" class="vector-tab-noicon mw-list-item"><a href="/w/index.php?title=Convolutional_neural_network&action=history" title="Past revisions of this page [h]" accesskey="h">View history</a></li> </ul> </div> </div> </nav> <nav class="vector-page-tools-landmark" aria-label="Page tools"> <div id="vector-page-tools-dropdown" class="vector-dropdown vector-page-tools-dropdown" > <input type="checkbox" id="vector-page-tools-dropdown-checkbox" role="button" aria-haspopup="true" data-event-name="ui.dropdown-vector-page-tools-dropdown" class="vector-dropdown-checkbox " aria-label="Tools" > <label id="vector-page-tools-dropdown-label" for="vector-page-tools-dropdown-checkbox" class="vector-dropdown-label cdx-button cdx-button--fake-button cdx-button--fake-button--enabled cdx-button--weight-quiet" aria-hidden="true" >Tools </label> <div class="vector-dropdown-content"> <div id="vector-page-tools-unpinned-container" class="vector-unpinned-container"> <div id="vector-page-tools" class="vector-page-tools vector-pinnable-element"> <div class="vector-pinnable-header vector-page-tools-pinnable-header vector-pinnable-header-unpinned" data-feature-name="page-tools-pinned" data-pinnable-element-id="vector-page-tools" data-pinned-container-id="vector-page-tools-pinned-container" data-unpinned-container-id="vector-page-tools-unpinned-container" > <div class="vector-pinnable-header-label">Tools</div> <button class="vector-pinnable-header-toggle-button vector-pinnable-header-pin-button" data-event-name="pinnable-header.vector-page-tools.pin">move to sidebar</button> <button class="vector-pinnable-header-toggle-button vector-pinnable-header-unpin-button" data-event-name="pinnable-header.vector-page-tools.unpin">hide</button> </div> <div id="p-cactions" class="vector-menu mw-portlet mw-portlet-cactions emptyPortlet vector-has-collapsible-items" title="More options" > <div class="vector-menu-heading"> Actions </div> <div class="vector-menu-content"> <ul class="vector-menu-content-list"> <li id="ca-more-view" class="selected vector-more-collapsible-item mw-list-item"><a href="/wiki/Convolutional_neural_network">Read</a></li><li id="ca-more-edit" class="vector-more-collapsible-item mw-list-item"><a href="/w/index.php?title=Convolutional_neural_network&action=edit" title="Edit this page [e]" accesskey="e">Edit</a></li><li id="ca-more-history" class="vector-more-collapsible-item mw-list-item"><a href="/w/index.php?title=Convolutional_neural_network&action=history">View history</a></li> </ul> </div> </div> <div id="p-tb" class="vector-menu mw-portlet mw-portlet-tb" > <div class="vector-menu-heading"> General </div> <div class="vector-menu-content"> <ul class="vector-menu-content-list"> <li id="t-whatlinkshere" class="mw-list-item"><a href="/wiki/Special:WhatLinksHere/Convolutional_neural_network" title="List of all English Wikipedia pages containing links to this page [j]" accesskey="j">What links here</a></li><li id="t-recentchangeslinked" class="mw-list-item"><a href="/wiki/Special:RecentChangesLinked/Convolutional_neural_network" rel="nofollow" title="Recent changes in pages linked from this page [k]" accesskey="k">Related changes</a></li><li id="t-upload" class="mw-list-item"><a href="/wiki/Wikipedia:File_Upload_Wizard" title="Upload files [u]" accesskey="u">Upload file</a></li><li id="t-specialpages" class="mw-list-item"><a href="/wiki/Special:SpecialPages" title="A list of all special pages [q]" accesskey="q">Special pages</a></li><li id="t-permalink" class="mw-list-item"><a href="/w/index.php?title=Convolutional_neural_network&oldid=1257441256" title="Permanent link to this revision of this page">Permanent link</a></li><li id="t-info" class="mw-list-item"><a href="/w/index.php?title=Convolutional_neural_network&action=info" title="More information about this page">Page information</a></li><li id="t-cite" class="mw-list-item"><a href="/w/index.php?title=Special:CiteThisPage&page=Convolutional_neural_network&id=1257441256&wpFormIdentifier=titleform" title="Information on how to cite this page">Cite this page</a></li><li id="t-urlshortener" class="mw-list-item"><a href="/w/index.php?title=Special:UrlShortener&url=https%3A%2F%2Fen.wikipedia.org%2Fwiki%2FConvolutional_neural_network">Get shortened URL</a></li><li id="t-urlshortener-qrcode" class="mw-list-item"><a href="/w/index.php?title=Special:QrCode&url=https%3A%2F%2Fen.wikipedia.org%2Fwiki%2FConvolutional_neural_network">Download QR code</a></li> </ul> </div> </div> <div id="p-coll-print_export" class="vector-menu mw-portlet mw-portlet-coll-print_export" > <div class="vector-menu-heading"> Print/export </div> <div class="vector-menu-content"> <ul class="vector-menu-content-list"> <li id="coll-download-as-rl" class="mw-list-item"><a href="/w/index.php?title=Special:DownloadAsPdf&page=Convolutional_neural_network&action=show-download-screen" title="Download this page as a PDF file">Download as PDF</a></li><li id="t-print" class="mw-list-item"><a href="/w/index.php?title=Convolutional_neural_network&printable=yes" title="Printable version of this page [p]" accesskey="p">Printable version</a></li> </ul> </div> </div> <div id="p-wikibase-otherprojects" class="vector-menu mw-portlet mw-portlet-wikibase-otherprojects" > <div class="vector-menu-heading"> In other projects </div> <div class="vector-menu-content"> <ul class="vector-menu-content-list"> <li id="t-wikibase" class="wb-otherproject-link wb-otherproject-wikibase-dataitem mw-list-item"><a href="https://www.wikidata.org/wiki/Special:EntityPage/Q17084460" title="Structured data on this page hosted by Wikidata [g]" accesskey="g">Wikidata item</a></li> </ul> </div> </div> </div> </div> </div> </div> </nav> </div> </div> </div> <div class="vector-column-end"> <div class="vector-sticky-pinned-container"> <nav class="vector-page-tools-landmark" aria-label="Page tools"> <div id="vector-page-tools-pinned-container" class="vector-pinned-container"> </div> </nav> <nav class="vector-appearance-landmark" aria-label="Appearance"> <div id="vector-appearance-pinned-container" class="vector-pinned-container"> <div id="vector-appearance" class="vector-appearance vector-pinnable-element"> <div class="vector-pinnable-header vector-appearance-pinnable-header vector-pinnable-header-pinned" data-feature-name="appearance-pinned" data-pinnable-element-id="vector-appearance" data-pinned-container-id="vector-appearance-pinned-container" data-unpinned-container-id="vector-appearance-unpinned-container" > <div class="vector-pinnable-header-label">Appearance</div> <button class="vector-pinnable-header-toggle-button vector-pinnable-header-pin-button" data-event-name="pinnable-header.vector-appearance.pin">move to sidebar</button> <button class="vector-pinnable-header-toggle-button vector-pinnable-header-unpin-button" data-event-name="pinnable-header.vector-appearance.unpin">hide</button> </div> </div> </div> </nav> </div> </div> <div id="bodyContent" class="vector-body" aria-labelledby="firstHeading" data-mw-ve-target-container> <div class="vector-body-before-content"> <div class="mw-indicators"> </div> <div id="siteSub" class="noprint">From Wikipedia, the free encyclopedia</div> </div> <div id="contentSub"><div id="mw-content-subtitle"></div></div> <div id="mw-content-text" class="mw-body-content"><div class="mw-content-ltr mw-parser-output" lang="en" dir="ltr"><div class="shortdescription nomobile noexcerpt noprint searchaux" style="display:none">Artificial neural network</div> <style data-mw-deduplicate="TemplateStyles:r1236090951">.mw-parser-output .hatnote{font-style:italic}.mw-parser-output div.hatnote{padding-left:1.6em;margin-bottom:0.5em}.mw-parser-output .hatnote i{font-style:normal}.mw-parser-output .hatnote+link+.hatnote{margin-top:-0.5em}@media print{body.ns-0 .mw-parser-output .hatnote{display:none!important}}</style><div role="note" class="hatnote navigation-not-searchable">For other uses, see <a href="/wiki/CNN_(disambiguation)" class="mw-disambig" title="CNN (disambiguation)">CNN (disambiguation)</a>.</div> <style data-mw-deduplicate="TemplateStyles:r1251242444">.mw-parser-output .ambox{border:1px solid #a2a9b1;border-left:10px solid #36c;background-color:#fbfbfb;box-sizing:border-box}.mw-parser-output .ambox+link+.ambox,.mw-parser-output .ambox+link+style+.ambox,.mw-parser-output .ambox+link+link+.ambox,.mw-parser-output .ambox+.mw-empty-elt+link+.ambox,.mw-parser-output .ambox+.mw-empty-elt+link+style+.ambox,.mw-parser-output .ambox+.mw-empty-elt+link+link+.ambox{margin-top:-1px}html body.mediawiki .mw-parser-output .ambox.mbox-small-left{margin:4px 1em 4px 0;overflow:hidden;width:238px;border-collapse:collapse;font-size:88%;line-height:1.25em}.mw-parser-output .ambox-speedy{border-left:10px solid #b32424;background-color:#fee7e6}.mw-parser-output .ambox-delete{border-left:10px solid #b32424}.mw-parser-output .ambox-content{border-left:10px solid #f28500}.mw-parser-output .ambox-style{border-left:10px solid #fc3}.mw-parser-output .ambox-move{border-left:10px solid #9932cc}.mw-parser-output .ambox-protection{border-left:10px solid #a2a9b1}.mw-parser-output .ambox .mbox-text{border:none;padding:0.25em 0.5em;width:100%}.mw-parser-output .ambox .mbox-image{border:none;padding:2px 0 2px 0.5em;text-align:center}.mw-parser-output .ambox .mbox-imageright{border:none;padding:2px 0.5em 2px 0;text-align:center}.mw-parser-output .ambox .mbox-empty-cell{border:none;padding:0;width:1px}.mw-parser-output .ambox .mbox-image-div{width:52px}@media(min-width:720px){.mw-parser-output .ambox{margin:0 10%}}@media print{body.ns-0 .mw-parser-output .ambox{display:none!important}}</style><table class="box-More_citations_needed plainlinks metadata ambox ambox-content ambox-Refimprove" role="presentation"><tbody><tr><td class="mbox-image"><div class="mbox-image-div"><a href="/wiki/File:Question_book-new.svg" class="mw-file-description"><img alt="" src="//upload.wikimedia.org/wikipedia/en/thumb/9/99/Question_book-new.svg/50px-Question_book-new.svg.png" decoding="async" width="50" height="39" class="mw-file-element" srcset="//upload.wikimedia.org/wikipedia/en/thumb/9/99/Question_book-new.svg/75px-Question_book-new.svg.png 1.5x, //upload.wikimedia.org/wikipedia/en/thumb/9/99/Question_book-new.svg/100px-Question_book-new.svg.png 2x" data-file-width="512" data-file-height="399" /></a></div></td><td class="mbox-text"><div class="mbox-text-span">This article needs additional citations for <a href="/wiki/Wikipedia:Verifiability" title="Wikipedia:Verifiability">verification</a>. Please help <a href="/wiki/Special:EditPage/Convolutional_neural_network" title="Special:EditPage/Convolutional neural network">improve this article</a> by <a href="/wiki/Help:Referencing_for_beginners" title="Help:Referencing for beginners">adding citations to reliable sources</a>. Unsourced material may be challenged and removed. Find sources: <a rel="nofollow" class="external text" href="https://www.google.com/search?as_eq=wikipedia&q=%22Convolutional+neural+network%22">"Convolutional neural network"</a> – <a rel="nofollow" class="external text" href="https://www.google.com/search?tbm=nws&q=%22Convolutional+neural+network%22+-wikipedia&tbs=ar:1">news</a> · <a rel="nofollow" class="external text" href="https://www.google.com/search?&q=%22Convolutional+neural+network%22&tbs=bkt:s&tbm=bks">newspapers</a> · <a rel="nofollow" class="external text" href="https://www.google.com/search?tbs=bks:1&q=%22Convolutional+neural+network%22+-wikipedia">books</a> · <a rel="nofollow" class="external text" href="https://scholar.google.com/scholar?q=%22Convolutional+neural+network%22">scholar</a> · <a rel="nofollow" class="external text" href="https://www.jstor.org/action/doBasicSearch?Query=%22Convolutional+neural+network%22&acc=on&wc=on">JSTOR</a> (June 2019) (<a href="/wiki/Help:Maintenance_template_removal" title="Help:Maintenance template removal">Learn how and when to remove this message</a>)</div></td></tr></tbody></table> <style data-mw-deduplicate="TemplateStyles:r1244144826">.mw-parser-output .machine-learning-list-title{background-color:#ddddff}html.skin-theme-clientpref-night .mw-parser-output .machine-learning-list-title{background-color:#222}@media(prefers-color-scheme:dark){html.skin-theme-clientpref-os .mw-parser-output .machine-learning-list-title{background-color:#222}}</style> <style data-mw-deduplicate="TemplateStyles:r1129693374">.mw-parser-output .hlist dl,.mw-parser-output .hlist ol,.mw-parser-output .hlist ul{margin:0;padding:0}.mw-parser-output .hlist dd,.mw-parser-output .hlist dt,.mw-parser-output .hlist li{margin:0;display:inline}.mw-parser-output .hlist.inline,.mw-parser-output .hlist.inline dl,.mw-parser-output .hlist.inline ol,.mw-parser-output .hlist.inline ul,.mw-parser-output .hlist dl dl,.mw-parser-output .hlist dl ol,.mw-parser-output .hlist dl ul,.mw-parser-output .hlist ol dl,.mw-parser-output .hlist ol ol,.mw-parser-output .hlist ol ul,.mw-parser-output .hlist ul dl,.mw-parser-output .hlist ul ol,.mw-parser-output .hlist ul ul{display:inline}.mw-parser-output .hlist .mw-empty-li{display:none}.mw-parser-output .hlist dt::after{content:": "}.mw-parser-output .hlist dd::after,.mw-parser-output .hlist li::after{content:" · ";font-weight:bold}.mw-parser-output .hlist dd:last-child::after,.mw-parser-output .hlist dt:last-child::after,.mw-parser-output .hlist li:last-child::after{content:none}.mw-parser-output .hlist dd dd:first-child::before,.mw-parser-output .hlist dd dt:first-child::before,.mw-parser-output .hlist dd li:first-child::before,.mw-parser-output .hlist dt dd:first-child::before,.mw-parser-output .hlist dt dt:first-child::before,.mw-parser-output .hlist dt li:first-child::before,.mw-parser-output .hlist li dd:first-child::before,.mw-parser-output .hlist li dt:first-child::before,.mw-parser-output .hlist li li:first-child::before{content:" (";font-weight:normal}.mw-parser-output .hlist dd dd:last-child::after,.mw-parser-output .hlist dd dt:last-child::after,.mw-parser-output .hlist dd li:last-child::after,.mw-parser-output .hlist dt dd:last-child::after,.mw-parser-output .hlist dt dt:last-child::after,.mw-parser-output .hlist dt li:last-child::after,.mw-parser-output .hlist li dd:last-child::after,.mw-parser-output .hlist li dt:last-child::after,.mw-parser-output .hlist li li:last-child::after{content:")";font-weight:normal}.mw-parser-output .hlist ol{counter-reset:listitem}.mw-parser-output .hlist ol>li{counter-increment:listitem}.mw-parser-output .hlist ol>li::before{content:" "counter(listitem)"\a0 "}.mw-parser-output .hlist dd ol>li:first-child::before,.mw-parser-output .hlist dt ol>li:first-child::before,.mw-parser-output .hlist li ol>li:first-child::before{content:" ("counter(listitem)"\a0 "}</style><style data-mw-deduplicate="TemplateStyles:r1246091330">.mw-parser-output .sidebar{width:22em;float:right;clear:right;margin:0.5em 0 1em 1em;background:var(--background-color-neutral-subtle,#f8f9fa);border:1px solid var(--border-color-base,#a2a9b1);padding:0.2em;text-align:center;line-height:1.4em;font-size:88%;border-collapse:collapse;display:table}body.skin-minerva .mw-parser-output .sidebar{display:table!important;float:right!important;margin:0.5em 0 1em 1em!important}.mw-parser-output .sidebar-subgroup{width:100%;margin:0;border-spacing:0}.mw-parser-output .sidebar-left{float:left;clear:left;margin:0.5em 1em 1em 0}.mw-parser-output .sidebar-none{float:none;clear:both;margin:0.5em 1em 1em 0}.mw-parser-output .sidebar-outer-title{padding:0 0.4em 0.2em;font-size:125%;line-height:1.2em;font-weight:bold}.mw-parser-output .sidebar-top-image{padding:0.4em}.mw-parser-output .sidebar-top-caption,.mw-parser-output .sidebar-pretitle-with-top-image,.mw-parser-output .sidebar-caption{padding:0.2em 0.4em 0;line-height:1.2em}.mw-parser-output .sidebar-pretitle{padding:0.4em 0.4em 0;line-height:1.2em}.mw-parser-output .sidebar-title,.mw-parser-output .sidebar-title-with-pretitle{padding:0.2em 0.8em;font-size:145%;line-height:1.2em}.mw-parser-output .sidebar-title-with-pretitle{padding:0.1em 0.4em}.mw-parser-output .sidebar-image{padding:0.2em 0.4em 0.4em}.mw-parser-output .sidebar-heading{padding:0.1em 0.4em}.mw-parser-output .sidebar-content{padding:0 0.5em 0.4em}.mw-parser-output .sidebar-content-with-subgroup{padding:0.1em 0.4em 0.2em}.mw-parser-output .sidebar-above,.mw-parser-output .sidebar-below{padding:0.3em 0.8em;font-weight:bold}.mw-parser-output .sidebar-collapse .sidebar-above,.mw-parser-output .sidebar-collapse .sidebar-below{border-top:1px solid #aaa;border-bottom:1px solid #aaa}.mw-parser-output .sidebar-navbar{text-align:right;font-size:115%;padding:0 0.4em 0.4em}.mw-parser-output .sidebar-list-title{padding:0 0.4em;text-align:left;font-weight:bold;line-height:1.6em;font-size:105%}.mw-parser-output .sidebar-list-title-c{padding:0 0.4em;text-align:center;margin:0 3.3em}@media(max-width:640px){body.mediawiki .mw-parser-output .sidebar{width:100%!important;clear:both;float:none!important;margin-left:0!important;margin-right:0!important}}body.skin--responsive .mw-parser-output .sidebar a>img{max-width:none!important}@media screen{html.skin-theme-clientpref-night .mw-parser-output .sidebar:not(.notheme) .sidebar-list-title,html.skin-theme-clientpref-night .mw-parser-output .sidebar:not(.notheme) .sidebar-title-with-pretitle{background:transparent!important}html.skin-theme-clientpref-night .mw-parser-output .sidebar:not(.notheme) .sidebar-title-with-pretitle a{color:var(--color-progressive)!important}}@media screen and (prefers-color-scheme:dark){html.skin-theme-clientpref-os .mw-parser-output .sidebar:not(.notheme) .sidebar-list-title,html.skin-theme-clientpref-os .mw-parser-output .sidebar:not(.notheme) .sidebar-title-with-pretitle{background:transparent!important}html.skin-theme-clientpref-os .mw-parser-output .sidebar:not(.notheme) .sidebar-title-with-pretitle a{color:var(--color-progressive)!important}}@media print{body.ns-0 .mw-parser-output .sidebar{display:none!important}}</style><style data-mw-deduplicate="TemplateStyles:r886047488">.mw-parser-output .nobold{font-weight:normal}</style><link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r886047488"><table class="sidebar sidebar-collapse nomobile nowraplinks"><tbody><tr><td class="sidebar-pretitle">Part of a series on</td></tr><tr><th class="sidebar-title-with-pretitle"><a href="/wiki/Machine_learning" title="Machine learning">Machine learning</a> and <a href="/wiki/Data_mining" title="Data mining">data mining</a></th></tr><tr><td class="sidebar-content"> <div class="sidebar-list mw-collapsible mw-collapsed machine-learning-list-title"><div class="sidebar-list-title" style="border-top:1px solid #aaa; text-align:center;;color: var(--color-base)">Paradigms</div><div class="sidebar-list-content mw-collapsible-content hlist"> <ul><li><a href="/wiki/Supervised_learning" title="Supervised learning">Supervised learning</a></li> <li><a href="/wiki/Unsupervised_learning" title="Unsupervised learning">Unsupervised learning</a></li> <li><a href="/wiki/Semi-supervised_learning" class="mw-redirect" title="Semi-supervised learning">Semi-supervised learning</a></li> <li><a href="/wiki/Self-supervised_learning" title="Self-supervised learning">Self-supervised learning</a></li> <li><a href="/wiki/Reinforcement_learning" title="Reinforcement learning">Reinforcement learning</a></li> <li><a href="/wiki/Meta-learning_(computer_science)" title="Meta-learning (computer science)">Meta-learning</a></li> <li><a href="/wiki/Online_machine_learning" title="Online machine learning">Online learning</a></li> <li><a href="/wiki/Batch_learning" class="mw-redirect" title="Batch learning">Batch learning</a></li> <li><a href="/wiki/Curriculum_learning" title="Curriculum learning">Curriculum learning</a></li> <li><a href="/wiki/Rule-based_machine_learning" title="Rule-based machine learning">Rule-based learning</a></li> <li><a href="/wiki/Neuro-symbolic_AI" title="Neuro-symbolic AI">Neuro-symbolic AI</a></li> <li><a href="/wiki/Neuromorphic_engineering" class="mw-redirect" title="Neuromorphic engineering">Neuromorphic engineering</a></li> <li><a href="/wiki/Quantum_machine_learning" title="Quantum machine learning">Quantum machine learning</a></li></ul></div></div></td> </tr><tr><td class="sidebar-content"> <div class="sidebar-list mw-collapsible mw-collapsed machine-learning-list-title"><div class="sidebar-list-title" style="border-top:1px solid #aaa; text-align:center;;color: var(--color-base)">Problems</div><div class="sidebar-list-content mw-collapsible-content hlist"> <ul><li><a href="/wiki/Statistical_classification" title="Statistical classification">Classification</a></li> <li><a href="/wiki/Generative_model" title="Generative model">Generative modeling</a></li> <li><a href="/wiki/Regression_analysis" title="Regression analysis">Regression</a></li> <li><a href="/wiki/Cluster_analysis" title="Cluster analysis">Clustering</a></li> <li><a href="/wiki/Dimensionality_reduction" title="Dimensionality reduction">Dimensionality reduction</a></li> <li><a href="/wiki/Density_estimation" title="Density estimation">Density estimation</a></li> <li><a href="/wiki/Anomaly_detection" title="Anomaly detection">Anomaly detection</a></li> <li><a href="/wiki/Data_cleaning" class="mw-redirect" title="Data cleaning">Data cleaning</a></li> <li><a href="/wiki/Automated_machine_learning" title="Automated machine learning">AutoML</a></li> <li><a href="/wiki/Association_rule_learning" title="Association rule learning">Association rules</a></li> <li><a href="/wiki/Semantic_analysis_(machine_learning)" title="Semantic analysis (machine learning)">Semantic analysis</a></li> <li><a href="/wiki/Structured_prediction" title="Structured prediction">Structured prediction</a></li> <li><a href="/wiki/Feature_engineering" title="Feature engineering">Feature engineering</a></li> <li><a href="/wiki/Feature_learning" title="Feature learning">Feature learning</a></li> <li><a href="/wiki/Learning_to_rank" title="Learning to rank">Learning to rank</a></li> <li><a href="/wiki/Grammar_induction" title="Grammar induction">Grammar induction</a></li> <li><a href="/wiki/Ontology_learning" title="Ontology learning">Ontology learning</a></li> <li><a href="/wiki/Multimodal_learning" title="Multimodal learning">Multimodal learning</a></li></ul></div></div></td> </tr><tr><td class="sidebar-content"> <div class="sidebar-list mw-collapsible mw-collapsed machine-learning-list-title"><div class="sidebar-list-title" style="border-top:1px solid #aaa; text-align:center;;color: var(--color-base)"><div style="display: inline-block; line-height: 1.2em; padding: .1em 0;"><a href="/wiki/Supervised_learning" title="Supervised learning">Supervised learning</a> (<a href="/wiki/Statistical_classification" title="Statistical classification">classification</a> • <a href="/wiki/Regression_analysis" title="Regression analysis">regression</a>) </div></div><div class="sidebar-list-content mw-collapsible-content hlist"> <ul><li><a href="/wiki/Apprenticeship_learning" title="Apprenticeship learning">Apprenticeship learning</a></li> <li><a href="/wiki/Decision_tree_learning" title="Decision tree learning">Decision trees</a></li> <li><a href="/wiki/Ensemble_learning" title="Ensemble learning">Ensembles</a> <ul><li><a href="/wiki/Bootstrap_aggregating" title="Bootstrap aggregating">Bagging</a></li> <li><a href="/wiki/Boosting_(machine_learning)" title="Boosting (machine learning)">Boosting</a></li> <li><a href="/wiki/Random_forest" title="Random forest">Random forest</a></li></ul></li> <li><a href="/wiki/K-nearest_neighbors_algorithm" title="K-nearest neighbors algorithm">k-NN</a></li> <li><a href="/wiki/Linear_regression" title="Linear regression">Linear regression</a></li> <li><a href="/wiki/Naive_Bayes_classifier" title="Naive Bayes classifier">Naive Bayes</a></li> <li><a href="/wiki/Artificial_neural_network" class="mw-redirect" title="Artificial neural network">Artificial neural networks</a></li> <li><a href="/wiki/Logistic_regression" title="Logistic regression">Logistic regression</a></li> <li><a href="/wiki/Perceptron" title="Perceptron">Perceptron</a></li> <li><a href="/wiki/Relevance_vector_machine" title="Relevance vector machine">Relevance vector machine (RVM)</a></li> <li><a href="/wiki/Support_vector_machine" title="Support vector machine">Support vector machine (SVM)</a></li></ul></div></div></td> </tr><tr><td class="sidebar-content"> <div class="sidebar-list mw-collapsible mw-collapsed machine-learning-list-title"><div class="sidebar-list-title" style="border-top:1px solid #aaa; text-align:center;;color: var(--color-base)"><a href="/wiki/Cluster_analysis" title="Cluster analysis">Clustering</a></div><div class="sidebar-list-content mw-collapsible-content hlist"> <ul><li><a href="/wiki/BIRCH" title="BIRCH">BIRCH</a></li> <li><a href="/wiki/CURE_algorithm" title="CURE algorithm">CURE</a></li> <li><a href="/wiki/Hierarchical_clustering" title="Hierarchical clustering">Hierarchical</a></li> <li><a href="/wiki/K-means_clustering" title="K-means clustering">k-means</a></li> <li><a href="/wiki/Fuzzy_clustering" title="Fuzzy clustering">Fuzzy</a></li> <li><a href="/wiki/Expectation%E2%80%93maximization_algorithm" title="Expectation–maximization algorithm">Expectation–maximization (EM)</a></li> <li> <a href="/wiki/DBSCAN" title="DBSCAN">DBSCAN</a></li> <li><a href="/wiki/OPTICS_algorithm" title="OPTICS algorithm">OPTICS</a></li> <li><a href="/wiki/Mean_shift" title="Mean shift">Mean shift</a></li></ul></div></div></td> </tr><tr><td class="sidebar-content"> <div class="sidebar-list mw-collapsible mw-collapsed machine-learning-list-title"><div class="sidebar-list-title" style="border-top:1px solid #aaa; text-align:center;;color: var(--color-base)"><a href="/wiki/Dimensionality_reduction" title="Dimensionality reduction">Dimensionality reduction</a></div><div class="sidebar-list-content mw-collapsible-content hlist"> <ul><li><a href="/wiki/Factor_analysis" title="Factor analysis">Factor analysis</a></li> <li><a href="/wiki/Canonical_correlation" title="Canonical correlation">CCA</a></li> <li><a href="/wiki/Independent_component_analysis" title="Independent component analysis">ICA</a></li> <li><a href="/wiki/Linear_discriminant_analysis" title="Linear discriminant analysis">LDA</a></li> <li><a href="/wiki/Non-negative_matrix_factorization" title="Non-negative matrix factorization">NMF</a></li> <li><a href="/wiki/Principal_component_analysis" title="Principal component analysis">PCA</a></li> <li><a href="/wiki/Proper_generalized_decomposition" title="Proper generalized decomposition">PGD</a></li> <li><a href="/wiki/T-distributed_stochastic_neighbor_embedding" title="T-distributed stochastic neighbor embedding">t-SNE</a></li> <li><a href="/wiki/Sparse_dictionary_learning" title="Sparse dictionary learning">SDL</a></li></ul></div></div></td> </tr><tr><td class="sidebar-content"> <div class="sidebar-list mw-collapsible mw-collapsed machine-learning-list-title"><div class="sidebar-list-title" style="border-top:1px solid #aaa; text-align:center;;color: var(--color-base)"><a href="/wiki/Structured_prediction" title="Structured prediction">Structured prediction</a></div><div class="sidebar-list-content mw-collapsible-content hlist"> <ul><li><a href="/wiki/Graphical_model" title="Graphical model">Graphical models</a> <ul><li><a href="/wiki/Bayesian_network" title="Bayesian network">Bayes net</a></li> <li><a href="/wiki/Conditional_random_field" title="Conditional random field">Conditional random field</a></li> <li><a href="/wiki/Hidden_Markov_model" title="Hidden Markov model">Hidden Markov</a></li></ul></li></ul></div></div></td> </tr><tr><td class="sidebar-content"> <div class="sidebar-list mw-collapsible mw-collapsed machine-learning-list-title"><div class="sidebar-list-title" style="border-top:1px solid #aaa; text-align:center;;color: var(--color-base)"><a href="/wiki/Anomaly_detection" title="Anomaly detection">Anomaly detection</a></div><div class="sidebar-list-content mw-collapsible-content hlist"> <ul><li><a href="/wiki/Random_sample_consensus" title="Random sample consensus">RANSAC</a></li> <li><a href="/wiki/K-nearest_neighbors_algorithm" title="K-nearest neighbors algorithm">k-NN</a></li> <li><a href="/wiki/Local_outlier_factor" title="Local outlier factor">Local outlier factor</a></li> <li><a href="/wiki/Isolation_forest" title="Isolation forest">Isolation forest</a></li></ul></div></div></td> </tr><tr><td class="sidebar-content"> <div class="sidebar-list mw-collapsible machine-learning-list-title"><div class="sidebar-list-title" style="border-top:1px solid #aaa; text-align:center;;color: var(--color-base)"><a href="/wiki/Artificial_neural_network" class="mw-redirect" title="Artificial neural network">Artificial neural network</a></div><div class="sidebar-list-content mw-collapsible-content hlist"> <ul><li><a href="/wiki/Autoencoder" title="Autoencoder">Autoencoder</a></li> <li><a href="/wiki/Deep_learning" title="Deep learning">Deep learning</a></li> <li><a href="/wiki/Feedforward_neural_network" title="Feedforward neural network">Feedforward neural network</a></li> <li><a href="/wiki/Recurrent_neural_network" title="Recurrent neural network">Recurrent neural network</a> <ul><li><a href="/wiki/Long_short-term_memory" title="Long short-term memory">LSTM</a></li> <li><a href="/wiki/Gated_recurrent_unit" title="Gated recurrent unit">GRU</a></li> <li><a href="/wiki/Echo_state_network" title="Echo state network">ESN</a></li> <li><a href="/wiki/Reservoir_computing" title="Reservoir computing">reservoir computing</a></li></ul></li> <li><a href="/wiki/Boltzmann_machine" title="Boltzmann machine">Boltzmann machine</a> <ul><li><a href="/wiki/Restricted_Boltzmann_machine" title="Restricted Boltzmann machine">Restricted</a></li></ul></li> <li><a href="/wiki/Generative_adversarial_network" title="Generative adversarial network">GAN</a></li> <li><a href="/wiki/Diffusion_model" title="Diffusion model">Diffusion model</a></li> <li><a href="/wiki/Self-organizing_map" title="Self-organizing map">SOM</a></li> <li><a class="mw-selflink selflink">Convolutional neural network</a> <ul><li><a href="/wiki/U-Net" title="U-Net">U-Net</a></li> <li><a href="/wiki/LeNet" title="LeNet">LeNet</a></li> <li><a href="/wiki/AlexNet" title="AlexNet">AlexNet</a></li> <li><a href="/wiki/DeepDream" title="DeepDream">DeepDream</a></li></ul></li> <li><a href="/wiki/Neural_radiance_field" title="Neural radiance field">Neural radiance field</a></li> <li><a href="/wiki/Transformer_(machine_learning_model)" class="mw-redirect" title="Transformer (machine learning model)">Transformer</a> <ul><li><a href="/wiki/Vision_transformer" title="Vision transformer">Vision</a></li></ul></li> <li><a href="/wiki/Mamba_(deep_learning_architecture)" title="Mamba (deep learning architecture)">Mamba</a></li> <li><a href="/wiki/Spiking_neural_network" title="Spiking neural network">Spiking neural network</a></li> <li><a href="/wiki/Memtransistor" title="Memtransistor">Memtransistor</a></li> <li><a href="/wiki/Electrochemical_RAM" title="Electrochemical RAM">Electrochemical RAM</a> (ECRAM)</li></ul></div></div></td> </tr><tr><td class="sidebar-content"> <div class="sidebar-list mw-collapsible mw-collapsed machine-learning-list-title"><div class="sidebar-list-title" style="border-top:1px solid #aaa; text-align:center;;color: var(--color-base)"><a href="/wiki/Reinforcement_learning" title="Reinforcement learning">Reinforcement learning</a></div><div class="sidebar-list-content mw-collapsible-content hlist"> <ul><li><a href="/wiki/Q-learning" title="Q-learning">Q-learning</a></li> <li><a href="/wiki/State%E2%80%93action%E2%80%93reward%E2%80%93state%E2%80%93action" title="State–action–reward–state–action">SARSA</a></li> <li><a href="/wiki/Temporal_difference_learning" title="Temporal difference learning">Temporal difference (TD)</a></li> <li><a href="/wiki/Multi-agent_reinforcement_learning" title="Multi-agent reinforcement learning">Multi-agent</a> <ul><li><a href="/wiki/Self-play_(reinforcement_learning_technique)" class="mw-redirect" title="Self-play (reinforcement learning technique)">Self-play</a></li></ul></li></ul></div></div></td> </tr><tr><td class="sidebar-content"> <div class="sidebar-list mw-collapsible mw-collapsed machine-learning-list-title"><div class="sidebar-list-title" style="border-top:1px solid #aaa; text-align:center;;color: var(--color-base)">Learning with humans</div><div class="sidebar-list-content mw-collapsible-content hlist"> <ul><li><a href="/wiki/Active_learning_(machine_learning)" title="Active learning (machine learning)">Active learning</a></li> <li><a href="/wiki/Crowdsourcing" title="Crowdsourcing">Crowdsourcing</a></li> <li><a href="/wiki/Human-in-the-loop" title="Human-in-the-loop">Human-in-the-loop</a></li> <li><a href="/wiki/Reinforcement_learning_from_human_feedback" title="Reinforcement learning from human feedback">RLHF</a></li></ul></div></div></td> </tr><tr><td class="sidebar-content"> <div class="sidebar-list mw-collapsible mw-collapsed machine-learning-list-title"><div class="sidebar-list-title" style="border-top:1px solid #aaa; text-align:center;;color: var(--color-base)">Model diagnostics</div><div class="sidebar-list-content mw-collapsible-content hlist"> <ul><li><a href="/wiki/Coefficient_of_determination" title="Coefficient of determination">Coefficient of determination</a></li> <li><a href="/wiki/Confusion_matrix" title="Confusion matrix">Confusion matrix</a></li> <li><a href="/wiki/Learning_curve_(machine_learning)" title="Learning curve (machine learning)">Learning curve</a></li> <li><a href="/wiki/Receiver_operating_characteristic" title="Receiver operating characteristic">ROC curve</a></li></ul></div></div></td> </tr><tr><td class="sidebar-content"> <div class="sidebar-list mw-collapsible mw-collapsed machine-learning-list-title"><div class="sidebar-list-title" style="border-top:1px solid #aaa; text-align:center;;color: var(--color-base)">Mathematical foundations</div><div class="sidebar-list-content mw-collapsible-content hlist"> <ul><li><a href="/wiki/Kernel_machines" class="mw-redirect" title="Kernel machines">Kernel machines</a></li> <li><a href="/wiki/Bias%E2%80%93variance_tradeoff" title="Bias–variance tradeoff">Bias–variance tradeoff</a></li> <li><a href="/wiki/Computational_learning_theory" title="Computational learning theory">Computational learning theory</a></li> <li><a href="/wiki/Empirical_risk_minimization" title="Empirical risk minimization">Empirical risk minimization</a></li> <li><a href="/wiki/Occam_learning" title="Occam learning">Occam learning</a></li> <li><a href="/wiki/Probably_approximately_correct_learning" title="Probably approximately correct learning">PAC learning</a></li> <li><a href="/wiki/Statistical_learning_theory" title="Statistical learning theory">Statistical learning</a></li> <li><a href="/wiki/Vapnik%E2%80%93Chervonenkis_theory" title="Vapnik–Chervonenkis theory">VC theory</a></li></ul></div></div></td> </tr><tr><td class="sidebar-content"> <div class="sidebar-list mw-collapsible mw-collapsed machine-learning-list-title"><div class="sidebar-list-title" style="border-top:1px solid #aaa; text-align:center;;color: var(--color-base)">Journals and conferences</div><div class="sidebar-list-content mw-collapsible-content hlist"> <ul><li><a href="/wiki/ECML_PKDD" title="ECML PKDD">ECML PKDD</a></li> <li><a href="/wiki/Conference_on_Neural_Information_Processing_Systems" title="Conference on Neural Information Processing Systems">NeurIPS</a></li> <li><a href="/wiki/International_Conference_on_Machine_Learning" title="International Conference on Machine Learning">ICML</a></li> <li><a href="/wiki/International_Conference_on_Learning_Representations" title="International Conference on Learning Representations">ICLR</a></li> <li><a href="/wiki/International_Joint_Conference_on_Artificial_Intelligence" title="International Joint Conference on Artificial Intelligence">IJCAI</a></li> <li><a href="/wiki/Machine_Learning_(journal)" title="Machine Learning (journal)">ML</a></li> <li><a href="/wiki/Journal_of_Machine_Learning_Research" title="Journal of Machine Learning Research">JMLR</a></li></ul></div></div></td> </tr><tr><td class="sidebar-content"> <div class="sidebar-list mw-collapsible mw-collapsed machine-learning-list-title"><div class="sidebar-list-title" style="border-top:1px solid #aaa; text-align:center;;color: var(--color-base)">Related articles</div><div class="sidebar-list-content mw-collapsible-content hlist"> <ul><li><a href="/wiki/Glossary_of_artificial_intelligence" title="Glossary of artificial intelligence">Glossary of artificial intelligence</a></li> <li><a href="/wiki/List_of_datasets_for_machine-learning_research" title="List of datasets for machine-learning research">List of datasets for machine-learning research</a> <ul><li><a href="/wiki/List_of_datasets_in_computer_vision_and_image_processing" title="List of datasets in computer vision and image processing">List of datasets in computer vision and image processing</a></li></ul></li> <li><a href="/wiki/Outline_of_machine_learning" title="Outline of machine learning">Outline of machine learning</a></li></ul></div></div></td> </tr><tr><td class="sidebar-navbar"><link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1129693374"><style data-mw-deduplicate="TemplateStyles:r1239400231">.mw-parser-output .navbar{display:inline;font-size:88%;font-weight:normal}.mw-parser-output .navbar-collapse{float:left;text-align:left}.mw-parser-output .navbar-boxtext{word-spacing:0}.mw-parser-output .navbar ul{display:inline-block;white-space:nowrap;line-height:inherit}.mw-parser-output .navbar-brackets::before{margin-right:-0.125em;content:"[ "}.mw-parser-output .navbar-brackets::after{margin-left:-0.125em;content:" ]"}.mw-parser-output .navbar li{word-spacing:-0.125em}.mw-parser-output .navbar a>span,.mw-parser-output .navbar a>abbr{text-decoration:inherit}.mw-parser-output .navbar-mini abbr{font-variant:small-caps;border-bottom:none;text-decoration:none;cursor:inherit}.mw-parser-output .navbar-ct-full{font-size:114%;margin:0 7em}.mw-parser-output .navbar-ct-mini{font-size:114%;margin:0 4em}html.skin-theme-clientpref-night .mw-parser-output .navbar li a abbr{color:var(--color-base)!important}@media(prefers-color-scheme:dark){html.skin-theme-clientpref-os .mw-parser-output .navbar li a abbr{color:var(--color-base)!important}}@media print{.mw-parser-output .navbar{display:none!important}}</style><div class="navbar plainlinks hlist navbar-mini"><ul><li class="nv-view"><a href="/wiki/Template:Machine_learning" title="Template:Machine learning"><abbr title="View this template">v</abbr></a></li><li class="nv-talk"><a href="/wiki/Template_talk:Machine_learning" title="Template talk:Machine learning"><abbr title="Discuss this template">t</abbr></a></li><li class="nv-edit"><a href="/wiki/Special:EditPage/Template:Machine_learning" title="Special:EditPage/Template:Machine learning"><abbr title="Edit this template">e</abbr></a></li></ul></div></td></tr></tbody></table> A convolutional neural network (CNN) is a <a href="/wiki/Regularization_(mathematics)" title="Regularization (mathematics)">regularized</a> type of <a href="/wiki/Feed-forward_neural_network" class="mw-redirect" title="Feed-forward neural network">feed-forward neural network</a> that learns <a href="/wiki/Feature_engineering" title="Feature engineering">features</a> by itself via <a href="/wiki/Filter_(signal_processing)" title="Filter (signal processing)">filter</a> (or kernel) optimization. This type of <a href="/wiki/Deep_learning" title="Deep learning">deep learning</a> network has been applied to process and make predictions from many different types of data including text, images and audio.<a href="#cite_note-1">[1]</a> Convolution-based networks are the de-facto standard in <a href="/wiki/Deep_learning" title="Deep learning">deep learning</a>-based approaches to <a href="/wiki/Computer_vision" title="Computer vision">computer vision</a> and image processing, and have only recently have been replaced -- in some cases -- by newer deep learning architectures such as the <a href="/wiki/Transformer_(deep_learning_architecture)" title="Transformer (deep learning architecture)">transformer</a>. <a href="/wiki/Vanishing_gradient_problem" title="Vanishing gradient problem">Vanishing gradients</a> and exploding gradients, seen during <a href="/wiki/Backpropagation" title="Backpropagation">backpropagation</a> in earlier neural networks, are prevented by using regularized weights over fewer connections.<a href="#cite_note-auto3-2">[2]</a><a href="#cite_note-auto2-3">[3]</a> For example, for each neuron in the fully-connected layer, 10,000 weights would be required for processing an image sized 100 × 100 pixels. However, applying cascaded convolution (or cross-correlation) kernels,<a href="#cite_note-4">[4]</a><a href="#cite_note-5">[5]</a> only 25 neurons are required to process 5x5-sized tiles.<a href="#cite_note-auto1-6">[6]</a><a href="#cite_note-homma-7">[7]</a> Higher-layer features are extracted from wider context windows, compared to lower-layer features. Some applications of CNNs include: <ul><li><a href="/wiki/Computer_vision" title="Computer vision">image and video recognition</a>,<a href="#cite_note-Valueva_Nagornov_Lyakhov_Valuev_2020_pp._232–243-8">[8]</a></li> <li><a href="/wiki/Recommender_system" title="Recommender system">recommender systems</a>,<a href="#cite_note-9">[9]</a></li> <li><a href="/wiki/Image_classification" class="mw-redirect" title="Image classification">image classification</a>,</li> <li><a href="/wiki/Image_segmentation" title="Image segmentation">image segmentation</a>,</li> <li><a href="/wiki/Medical_image_computing" title="Medical image computing">medical image analysis</a>,</li> <li><a href="/wiki/Natural_language_processing" title="Natural language processing">natural language processing</a>,<a href="#cite_note-10">[10]</a></li> <li><a href="/wiki/Brain%E2%80%93computer_interface" title="Brain–computer interface">brain–computer interfaces</a>,<a href="#cite_note-11">[11]</a> and</li> <li>financial <a href="/wiki/Time_series" title="Time series">time series</a>.<a href="#cite_note-Tsantekidis_7–12-12">[12]</a></li></ul> CNNs are also known as shift invariant or space invariant artificial neural networks, based on the shared-weight architecture of the <a href="/wiki/Convolution" title="Convolution">convolution</a> kernels or filters that slide along input features and provide translation-<a href="/wiki/Equivariant_map" title="Equivariant map">equivariant</a> responses known as feature maps.<a href="#cite_note-:0-13">[13]</a><a href="#cite_note-:1-14">[14]</a> Counter-intuitively, most convolutional neural networks are not <a href="/wiki/Translation_invariant" class="mw-redirect" title="Translation invariant">invariant to translation</a>, due to the downsampling operation they apply to the input.<a href="#cite_note-:6-15">[15]</a> <a href="/wiki/Feed-forward_neural_network" class="mw-redirect" title="Feed-forward neural network">Feed-forward neural networks</a> are usually fully connected networks, that is, each neuron in one <a href="/wiki/Layer_(deep_learning)" title="Layer (deep learning)">layer</a> is connected to all neurons in the next <a href="/wiki/Layer_(deep_learning)" title="Layer (deep learning)">layer</a>. The "full connectivity" of these networks makes them prone to <a href="/wiki/Overfitting" title="Overfitting">overfitting</a> data. Typical ways of regularization, or preventing overfitting, include: penalizing parameters during training (such as weight decay) or trimming connectivity (skipped connections, dropout, etc.) Robust datasets also increase the probability that CNNs will learn the generalized principles that characterize a given dataset rather than the biases of a poorly-populated set.<a href="#cite_note-16">[16]</a> Convolutional networks were <a href="/wiki/Mathematical_biology" class="mw-redirect" title="Mathematical biology">inspired</a> by <a href="/wiki/Biological" class="mw-redirect" title="Biological">biological</a> processes<a href="#cite_note-fukuneoscholar-17">[17]</a><a href="#cite_note-hubelwiesel1968-18">[18]</a><a href="#cite_note-intro-19">[19]</a><a href="#cite_note-robust_face_detection-20">[20]</a> in that the connectivity pattern between <a href="/wiki/Artificial_neuron" title="Artificial neuron">neurons</a> resembles the organization of the animal <a href="/wiki/Visual_cortex" title="Visual cortex">visual cortex</a>. Individual <a href="/wiki/Cortical_neuron" class="mw-redirect" title="Cortical neuron">cortical neurons</a> respond to stimuli only in a restricted region of the <a href="/wiki/Visual_field" title="Visual field">visual field</a> known as the <a href="/wiki/Receptive_field" title="Receptive field">receptive field</a>. The receptive fields of different neurons partially overlap such that they cover the entire visual field. CNNs use relatively little pre-processing compared to other <a href="/wiki/Image_classification" class="mw-redirect" title="Image classification">image classification algorithms</a>. This means that the network learns to optimize the <a href="/wiki/Filter_(signal_processing)" title="Filter (signal processing)">filters</a> (or kernels) through automated learning, whereas in traditional algorithms these filters are <a href="/wiki/Feature_engineering" title="Feature engineering">hand-engineered</a>. This independence from prior knowledge and human intervention in feature extraction is a major advantage.[<a href="/wiki/Wikipedia:Avoid_weasel_words" class="mw-redirect" title="Wikipedia:Avoid weasel words">to whom?</a>] <style data-mw-deduplicate="TemplateStyles:r886046785">.mw-parser-output .toclimit-2 .toclevel-1 ul,.mw-parser-output .toclimit-3 .toclevel-2 ul,.mw-parser-output .toclimit-4 .toclevel-3 ul,.mw-parser-output .toclimit-5 .toclevel-4 ul,.mw-parser-output .toclimit-6 .toclevel-5 ul,.mw-parser-output .toclimit-7 .toclevel-6 ul{display:none}</style><div class="toclimit-3"><meta property="mw:PageProp/toc" /></div> <div class="mw-heading mw-heading2"><h2 id="Architecture">Architecture</h2>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=1" title="Edit section: Architecture">edit</a>]</div> <figure typeof="mw:File/Thumb"><a href="/wiki/File:Comparison_image_neural_networks.svg" class="mw-file-description"><img src="//upload.wikimedia.org/wikipedia/commons/thumb/c/cc/Comparison_image_neural_networks.svg/480px-Comparison_image_neural_networks.svg.png" decoding="async" width="480" height="360" class="mw-file-element" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/c/cc/Comparison_image_neural_networks.svg/720px-Comparison_image_neural_networks.svg.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/c/cc/Comparison_image_neural_networks.svg/960px-Comparison_image_neural_networks.svg.png 2x" data-file-width="512" data-file-height="384" /></a><figcaption>Comparison of the <a href="/wiki/LeNet" title="LeNet">LeNet</a> and <a href="/wiki/AlexNet" title="AlexNet">AlexNet</a> convolution, pooling and dense layers (AlexNet image size should be 227×227×3, instead of 224×224×3, so the math will come out right. The original paper said different numbers, but Andrej Karpathy, the head of computer vision at Tesla, said it should be 227×227×3 (he said Alex did not describe why he put 224×224×3). The next convolution should be 11×11 with stride 4: 55×55×96 (instead of 54×54×96). It would be calculated, for example, as: [(input width 227 - kernel width 11) / stride 4] + 1 = [(227 - 11) / 4] + 1 = 55. Since the kernel output is the same length as width, its area is 55×55.)</figcaption></figure> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1236090951"><div role="note" class="hatnote navigation-not-searchable">Main article: <a href="/wiki/Layer_(deep_learning)" title="Layer (deep learning)">Layer (deep learning)</a></div> A convolutional neural network consists of an input layer, <a href="/wiki/Artificial_neural_network#Organization" class="mw-redirect" title="Artificial neural network">hidden layers</a> and an output layer. In a convolutional neural network, the hidden layers include one or more layers that perform convolutions. Typically this includes a layer that performs a <a href="/wiki/Dot_product" title="Dot product">dot product</a> of the convolution kernel with the layer's input matrix. This product is usually the <a href="/wiki/Frobenius_inner_product" title="Frobenius inner product">Frobenius inner product</a>, and its activation function is commonly <a href="/wiki/Rectifier_(neural_networks)" title="Rectifier (neural networks)">ReLU</a>. As the convolution kernel slides along the input matrix for the layer, the convolution operation generates a feature map, which in turn contributes to the input of the next layer. This is followed by other layers such as <a href="/wiki/Pooling_layer" title="Pooling layer">pooling layers</a>, fully connected layers, and normalization layers. Here it should be noted how close a convolutional neural network is to a <a href="/wiki/Matched_filter" title="Matched filter">matched filter</a>.<a href="#cite_note-21">[21]</a> <div class="mw-heading mw-heading3"><h3 id="Convolutional_layers">Convolutional layers</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=2" title="Edit section: Convolutional layers">edit</a>]</div> In a CNN, the input is a <a href="/wiki/Tensor_(machine_learning)" title="Tensor (machine learning)">tensor</a> with shape: (number of inputs) × (input height) × (input width) × (input <a href="/wiki/Channel_(digital_image)" title="Channel (digital image)">channels</a>) After passing through a convolutional layer, the image becomes abstracted to a feature map, also called an activation map, with shape: (number of inputs) × (feature map height) × (feature map width) × (feature map <a href="/wiki/Channel_(digital_image)" title="Channel (digital image)">channels</a>). Convolutional layers convolve the input and pass its result to the next layer. This is similar to the response of a neuron in the visual cortex to a specific stimulus.<a href="#cite_note-deeplearning-22">[22]</a> Each convolutional neuron processes data only for its <a href="/wiki/Receptive_field" title="Receptive field">receptive field</a>. <figure class="mw-default-size" typeof="mw:File/Thumb"><a href="/wiki/File:1D_Convolutional_Neural_Network_feed_forward_example.png" class="mw-file-description"><img src="//upload.wikimedia.org/wikipedia/commons/thumb/3/31/1D_Convolutional_Neural_Network_feed_forward_example.png/220px-1D_Convolutional_Neural_Network_feed_forward_example.png" decoding="async" width="220" height="100" class="mw-file-element" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/3/31/1D_Convolutional_Neural_Network_feed_forward_example.png/330px-1D_Convolutional_Neural_Network_feed_forward_example.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/3/31/1D_Convolutional_Neural_Network_feed_forward_example.png/440px-1D_Convolutional_Neural_Network_feed_forward_example.png 2x" data-file-width="661" data-file-height="301" /></a><figcaption>1D convolutional neural network feed forward example</figcaption></figure> Although <a href="/wiki/Multilayer_perceptron" title="Multilayer perceptron">fully connected feedforward neural networks</a> can be used to learn features and classify data, this architecture is generally impractical for larger inputs (e.g., high-resolution images), which would require massive numbers of neurons because each pixel is a relevant input feature. A fully connected layer for an image of size 100 × 100 has 10,000 weights for each neuron in the second layer. Convolution reduces the number of free parameters, allowing the network to be deeper.<a href="#cite_note-auto1-6">[6]</a> For example, using a 5 × 5 tiling region, each with the same shared weights, requires only 25 neurons. Using regularized weights over fewer parameters avoids the vanishing gradients and exploding gradients problems seen during <a href="/wiki/Backpropagation" title="Backpropagation">backpropagation</a> in earlier neural networks.<a href="#cite_note-auto3-2">[2]</a><a href="#cite_note-auto2-3">[3]</a> To speed processing, standard convolutional layers can be replaced by depthwise separable convolutional layers,<a href="#cite_note-23">[23]</a> which are based on a depthwise convolution followed by a pointwise convolution. The depthwise convolution is a spatial convolution applied independently over each channel of the input tensor, while the pointwise convolution is a standard convolution restricted to the use of <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle 1\times 1}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mn>1</mn> <mo>×</mo> <mn>1</mn> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle 1\times 1}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/2b4bf91a527dc01af9ef6ace81199becf1308e00" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.338ex; width:5.165ex; height:2.176ex;" alt="{\displaystyle 1\times 1}"> kernels. <div class="mw-heading mw-heading3"><h3 id="Pooling_layers">Pooling layers</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=3" title="Edit section: Pooling layers">edit</a>]</div> Convolutional networks may include local and/or global pooling layers along with traditional convolutional layers. Pooling layers reduce the dimensions of data by combining the outputs of neuron clusters at one layer into a single neuron in the next layer. Local pooling combines small clusters, tiling sizes such as 2 × 2 are commonly used. Global pooling acts on all the neurons of the feature map.<a href="#cite_note-flexible-24">[24]</a><a href="#cite_note-25">[25]</a> There are two common types of pooling in popular use: max and average. Max pooling uses the maximum value of each local cluster of neurons in the feature map,<a href="#cite_note-Yamaguchi111990-26">[26]</a><a href="#cite_note-mcdns-27">[27]</a> while average pooling takes the average value. <div class="mw-heading mw-heading3"><h3 id="Fully_connected_layers">Fully connected layers</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=4" title="Edit section: Fully connected layers">edit</a>]</div> Fully connected layers connect every neuron in one layer to every neuron in another layer. It is the same as a traditional <a href="/wiki/Multilayer_perceptron" title="Multilayer perceptron">multilayer perceptron</a> neural network (MLP). The flattened matrix goes through a fully connected layer to classify the images. <div class="mw-heading mw-heading3"><h3 id="Receptive_field">Receptive field</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=5" title="Edit section: Receptive field">edit</a>]</div> In neural networks, each neuron receives input from some number of locations in the previous layer. In a convolutional layer, each neuron receives input from only a restricted area of the previous layer called the neuron's receptive field. Typically the area is a square (e.g. 5 by 5 neurons). Whereas, in a fully connected layer, the receptive field is the entire previous layer. Thus, in each convolutional layer, each neuron takes input from a larger area in the input than previous layers. This is due to applying the convolution over and over, which takes the value of a pixel into account, as well as its surrounding pixels. When using dilated layers, the number of pixels in the receptive field remains constant, but the field is more sparsely populated as its dimensions grow when combining the effect of several layers. To manipulate the receptive field size as desired, there are some alternatives to the standard convolutional layer. For example, atrous or dilated convolution<a href="#cite_note-28">[28]</a><a href="#cite_note-29">[29]</a> expands the receptive field size without increasing the number of parameters by interleaving visible and blind regions. Moreover, a single dilated convolutional layer can comprise filters with multiple dilation ratios,<a href="#cite_note-30">[30]</a> thus having a variable receptive field size. <div class="mw-heading mw-heading3"><h3 id="Weights">Weights</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=6" title="Edit section: Weights">edit</a>]</div> Each neuron in a neural network computes an output value by applying a specific function to the input values received from the receptive field in the previous layer. The function that is applied to the input values is determined by a vector of weights and a bias (typically real numbers). Learning consists of iteratively adjusting these biases and weights. The vectors of weights and biases are called filters and represent particular <a href="/wiki/Feature_(machine_learning)" title="Feature (machine learning)">features</a> of the input (e.g., a particular shape). A distinguishing feature of CNNs is that many neurons can share the same filter. This reduces the <a href="/wiki/Memory_footprint" title="Memory footprint">memory footprint</a> because a single bias and a single vector of weights are used across all receptive fields that share that filter, as opposed to each receptive field having its own bias and vector weighting.<a href="#cite_note-LeCun-31">[31]</a> <div class="mw-heading mw-heading3"><h3 id="Deconvolutional">Deconvolutional</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=7" title="Edit section: Deconvolutional">edit</a>]</div> A deconvolutional neural network is essentially the reverse of a CNN. It consists of deconvolutional layers and unpooling layers.<a href="#cite_note-32">[32]</a> A deconvolutional layer is the transpose of a convolutional layer. Specifically, a convolutional layer can be written as a multiplication with a matrix, and a deconvolutional layer is multiplication with the transpose of that matrix.<a href="#cite_note-33">[33]</a> An unpooling layer expands the layer. The max-unpooling layer is the simplest, as it simply copies each entry multiple times. For example, a 2-by-2 max-unpooling layer is <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle [x]\mapsto {\begin{bmatrix}x&x\\x&x\end{bmatrix}}}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mo stretchy="false">[</mo> <mi>x</mi> <mo stretchy="false">]</mo> <mo stretchy="false">↦</mo> <mrow class="MJX-TeXAtom-ORD"> <mrow> <mo>[</mo> <mtable rowspacing="4pt" columnspacing="1em"> <mtr> <mtd> <mi>x</mi> </mtd> <mtd> <mi>x</mi> </mtd> </mtr> <mtr> <mtd> <mi>x</mi> </mtd> <mtd> <mi>x</mi> </mtd> </mtr> </mtable> <mo>]</mo> </mrow> </mrow> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle [x]\mapsto {\begin{bmatrix}x&x\\x&x\end{bmatrix}}}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/ba907f707b81817e69c003905058b928e9097b86" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -2.505ex; width:14.426ex; height:6.176ex;" alt="{\displaystyle [x]\mapsto {\begin{bmatrix}x&x\\x&x\end{bmatrix}}}">. Deconvolution layers are used in image generators. By default, it creates periodic checkerboard artifact, which can be fixed by upscale-then-convolve.<a href="#cite_note-34">[34]</a> <div class="mw-heading mw-heading2"><h2 id="History">History</h2>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=8" title="Edit section: History">edit</a>]</div> CNN are often compared to the way the brain achieves vision processing in living <a href="/wiki/Organisms" class="mw-redirect" title="Organisms">organisms</a>.<a href="#cite_note-35">[35]</a> <div class="mw-heading mw-heading3"><h3 id="Receptive_fields_in_the_visual_cortex">Receptive fields in the visual cortex</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=9" title="Edit section: Receptive fields in the visual cortex">edit</a>]</div> Work by <a href="/wiki/David_H._Hubel" title="David H. Hubel">Hubel</a> and <a href="/wiki/Torsten_Wiesel" title="Torsten Wiesel">Wiesel</a> in the 1950s and 1960s showed that cat <a href="/wiki/Visual_cortex" title="Visual cortex">visual cortices</a> contain neurons that individually respond to small regions of the <a href="/wiki/Visual_field" title="Visual field">visual field</a>. Provided the eyes are not moving, the region of visual space within which visual stimuli affect the firing of a single neuron is known as its <a href="/wiki/Receptive_field" title="Receptive field">receptive field</a>.<a href="#cite_note-:4-36">[36]</a> Neighboring cells have similar and overlapping receptive fields. Receptive field size and location varies systematically across the cortex to form a complete map of visual space.[<a href="/wiki/Wikipedia:Citation_needed" title="Wikipedia:Citation needed">citation needed</a>] The cortex in each hemisphere represents the contralateral <a href="/wiki/Visual_field" title="Visual field">visual field</a>.[<a href="/wiki/Wikipedia:Citation_needed" title="Wikipedia:Citation needed">citation needed</a>] Their 1968 paper identified two basic visual cell types in the brain:<a href="#cite_note-hubelwiesel1968-18">[18]</a> <ul><li><a href="/wiki/Simple_cell" title="Simple cell">simple cells</a>, whose output is maximized by straight edges having particular orientations within their receptive field</li> <li><a href="/wiki/Complex_cell" title="Complex cell">complex cells</a>, which have larger <a href="/wiki/Receptive_field" title="Receptive field">receptive fields</a>, whose output is insensitive to the exact position of the edges in the field.</li></ul> Hubel and Wiesel also proposed a cascading model of these two types of cells for use in pattern recognition tasks.<a href="#cite_note-37">[37]</a><a href="#cite_note-:4-36">[36]</a> <div class="mw-heading mw-heading3"><h3 id="Neocognitron,_origin_of_the_CNN_architecture">Neocognitron, origin of the CNN architecture</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=10" title="Edit section: Neocognitron, origin of the CNN architecture">edit</a>]</div> Inspired by Hubel and Wiesel's work, in 1969, <a href="/wiki/Kunihiko_Fukushima" title="Kunihiko Fukushima">Kunihiko Fukushima</a> published a deep CNN that uses <a href="/wiki/Rectifier_(neural_networks)" title="Rectifier (neural networks)">ReLU</a> <a href="/wiki/Activation_function" title="Activation function">activation function</a>.<a href="#cite_note-Fukushima1969-38">[38]</a> Unlike most modern networks, this network used hand-designed kernels. It was not used in his neocognitron, since all the weights were nonnegative; lateral inhibition was used instead. The rectifier has become the most popular activation function for CNNs and <a href="/wiki/Deep_learning" title="Deep learning">deep neural networks</a> in general.<a href="#cite_note-39">[39]</a> The "<a href="/wiki/Neocognitron" title="Neocognitron">neocognitron</a>" was introduced by <a href="/wiki/Kunihiko_Fukushima" title="Kunihiko Fukushima">Kunihiko Fukushima</a> in 1979.<a href="#cite_note-40">[40]</a><a href="#cite_note-intro-19">[19]</a><a href="#cite_note-fukuneoscholar-17">[17]</a> The kernels were trained by <a href="/wiki/Unsupervised_learning" title="Unsupervised learning">unsupervised learning</a>. It was inspired by the above-mentioned work of Hubel and Wiesel. The neocognitron introduced the two basic types of layers: <ul><li>"S-layer": a shared-weights receptive-field layer, later known as a convolutional layer, which contains units whose receptive fields cover a patch of the previous layer. A shared-weights receptive-field group (a "plane" in neocognitron terminology) is often called a filter, and a layer typically has several such filters.</li> <li>"C-layer": a downsampling layer that contain units whose receptive fields cover patches of previous convolutional layers. Such a unit typically computes a weighted average of the activations of the units in its patch, and applies inhibition (divisive normalization) pooled from a somewhat larger patch and across different filters in a layer, and applies a saturating activation function. The patch weights are nonnegative and are not trainable in the original neocognitron. The downsampling and competitive inhibition help to classify features and objects in visual scenes even when the objects are shifted.</li></ul> In a variant of the neocognitron called the cresceptron, instead of using Fukushima's spatial averaging with inhibition and saturation, J. Weng et al. in 1993 introduced a method called max-pooling where a downsampling unit computes the maximum of the activations of the units in its patch.<a href="#cite_note-weng1993-41">[41]</a> Max-pooling is often used in modern CNNs.<a href="#cite_note-schdeepscholar-42">[42]</a> Several <a href="/wiki/Supervised_learning" title="Supervised learning">supervised</a> and <a href="/wiki/Unsupervised_learning" title="Unsupervised learning">unsupervised learning</a> algorithms have been proposed over the decades to train the weights of a neocognitron.<a href="#cite_note-fukuneoscholar-17">[17]</a> Today, however, the CNN architecture is usually trained through <a href="/wiki/Backpropagation" title="Backpropagation">backpropagation</a>. <div class="mw-heading mw-heading3"><h3 id="Convolution_in_time">Convolution in time</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=11" title="Edit section: Convolution in time">edit</a>]</div> The term "convolution" first appears in neural networks in a paper by Toshiteru Homma, Les Atlas, and Robert Marks II at the first <a href="/wiki/Conference_on_Neural_Information_Processing_Systems" title="Conference on Neural Information Processing Systems">Conference on Neural Information Processing Systems</a> in 1987. Their paper replaced multiplication with convolution in time, inherently providing shift invariance, motivated by and connecting more directly to the <a href="/wiki/Linear_shift-invariant_filter" class="mw-redirect" title="Linear shift-invariant filter">signal-processing concept of a filter</a>, and demonstrated it on a speech recognition task.<a href="#cite_note-homma-7">[7]</a> They also pointed out that as a data-trainable system, convolution is essentially equivalent to correlation since reversal of the weights does not affect the final learned function ("For convenience, we denote * as correlation instead of convolution. Note that convolving a(t) with b(t) is equivalent to correlating a(-t) with b(t).").<a href="#cite_note-homma-7">[7]</a> Modern CNN implementations typically do correlation and call it convolution, for convenience, as they did here. <div class="mw-heading mw-heading3"><h3 id="Time_delay_neural_networks">Time delay neural networks</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=12" title="Edit section: Time delay neural networks">edit</a>]</div> The <a href="/wiki/Time_delay_neural_network" title="Time delay neural network">time delay neural network</a> (TDNN) was introduced in 1987 by <a href="/wiki/Alex_Waibel" title="Alex Waibel">Alex Waibel</a> et al. for phoneme recognition and was one of the first convolutional networks, as it achieved shift-invariance.<a href="#cite_note-Waibel1987-43">[43]</a> A TDNN is a 1-D convolutional neural net where the convolution is performed along the time axis of the data. It is the first CNN utilizing weight sharing in combination with a training by gradient descent, using <a href="/wiki/Backpropagation" title="Backpropagation">backpropagation</a>.<a href="#cite_note-speechsignal-44">[44]</a> Thus, while also using a pyramidal structure as in the neocognitron, it performed a global optimization of the weights instead of a local one.<a href="#cite_note-Waibel1987-43">[43]</a> TDNNs are convolutional networks that share weights along the temporal dimension.<a href="#cite_note-45">[45]</a> They allow speech signals to be processed time-invariantly. In 1990 Hampshire and Waibel introduced a variant that performs a two-dimensional convolution.<a href="#cite_note-Hampshire1990-46">[46]</a> Since these TDNNs operated on spectrograms, the resulting phoneme recognition system was invariant to both time and frequency shifts, as with images processed by a neocognitron. TDNNs improved the performance of far-distance speech recognition.<a href="#cite_note-Ko2017-47">[47]</a> <div class="mw-heading mw-heading3"><h3 id="Image_recognition_with_CNNs_trained_by_gradient_descent">Image recognition with CNNs trained by gradient descent</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=13" title="Edit section: Image recognition with CNNs trained by gradient descent">edit</a>]</div> Denker et al. (1989) designed a 2-D CNN system to recognize hand-written <a href="/wiki/ZIP_Code" title="ZIP Code">ZIP Code</a> numbers.<a href="#cite_note-48">[48]</a> However, the lack of an efficient training method to determine the kernel coefficients of the involved convolutions meant that all the coefficients had to be laboriously hand-designed.<a href="#cite_note-:2-49">[49]</a> Following the advances in the training of 1-D CNNs by Waibel et al. (1987), <a href="/wiki/Yann_LeCun" title="Yann LeCun">Yann LeCun</a> et al. (1989)<a href="#cite_note-:2-49">[49]</a> used back-propagation to learn the convolution kernel coefficients directly from images of hand-written numbers. Learning was thus fully automatic, performed better than manual coefficient design, and was suited to a broader range of image recognition problems and image types. Wei Zhang et al. (1988)<a href="#cite_note-:0-13">[13]</a><a href="#cite_note-:1-14">[14]</a> used back-propagation to train the convolution kernels of a CNN for alphabets recognition. The model was called shift-invariant pattern recognition neural network before the name CNN was coined later in the early 1990s. Wei Zhang et al. also applied the same CNN without the last fully connected layer for medical image object segmentation (1991)<a href="#cite_note-:wz1991-50">[50]</a> and breast cancer detection in mammograms (1994).<a href="#cite_note-:wz1994-51">[51]</a> This approach became a foundation of modern <a href="/wiki/Computer_vision" title="Computer vision">computer vision</a>. <div class="mw-heading mw-heading4"><h4 id="Max_pooling">Max pooling</h4>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=14" title="Edit section: Max pooling">edit</a>]</div> In 1990 Yamaguchi et al. introduced the concept of max pooling, a fixed filtering operation that calculates and propagates the maximum value of a given region. They did so by combining TDNNs with max pooling to realize a speaker-independent isolated word recognition system.<a href="#cite_note-Yamaguchi111990-26">[26]</a> In their system they used several TDNNs per word, one for each <a href="/wiki/Syllable" title="Syllable">syllable</a>. The results of each TDNN over the input signal were combined using max pooling and the outputs of the pooling layers were then passed on to networks performing the actual word classification. <div class="mw-heading mw-heading4"><h4 id="LeNet-5">LeNet-5</h4>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=15" title="Edit section: LeNet-5">edit</a>]</div> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1236090951"><div role="note" class="hatnote navigation-not-searchable">Main article: <a href="/wiki/LeNet" title="LeNet">LeNet</a></div> LeNet-5, a pioneering 7-level convolutional network by <a href="/wiki/Yann_LeCun" title="Yann LeCun">LeCun</a> et al. in 1995,<a href="#cite_note-lecun95-52">[52]</a> classifies hand-written numbers on checks (<a href="/wiki/British_English_language" class="mw-redirect" title="British English language">British English</a>: cheques) digitized in 32x32 pixel images. The ability to process higher-resolution images requires larger and more layers of convolutional neural networks, so this technique is constrained by the availability of computing resources. It was superior than other commercial courtesy amount reading systems (as of 1995). The system was integrated in <a href="/wiki/NCR_Voyix" title="NCR Voyix">NCR</a>'s check reading systems, and fielded in several American banks since June 1996, reading millions of checks per day.<a href="#cite_note-53">[53]</a> <div class="mw-heading mw-heading3"><h3 id="Shift-invariant_neural_network">Shift-invariant neural network</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=16" title="Edit section: Shift-invariant neural network">edit</a>]</div> A shift-invariant neural network was proposed by Wei Zhang et al. for image character recognition in 1988.<a href="#cite_note-:0-13">[13]</a><a href="#cite_note-:1-14">[14]</a> It is a modified Neocognitron by keeping only the convolutional interconnections between the image feature layers and the last fully connected layer. The model was trained with back-propagation. The training algorithm was further improved in 1991<a href="#cite_note-54">[54]</a> to improve its generalization ability. The model architecture was modified by removing the last fully connected layer and applied for medical image segmentation (1991)<a href="#cite_note-:wz1991-50">[50]</a> and automatic detection of breast cancer in <a href="/wiki/Mammography" title="Mammography">mammograms (1994)</a>.<a href="#cite_note-:wz1994-51">[51]</a> A different convolution-based design was proposed in 1988<a href="#cite_note-55">[55]</a> for application to decomposition of one-dimensional <a href="/wiki/Electromyography" title="Electromyography">electromyography</a> convolved signals via de-convolution. This design was modified in 1989 to other de-convolution-based designs.<a href="#cite_note-56">[56]</a><a href="#cite_note-57">[57]</a> <div class="mw-heading mw-heading3"><h3 id="GPU_implementations">GPU implementations</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=17" title="Edit section: GPU implementations">edit</a>]</div> Although CNNs were invented in the 1980s, their breakthrough in the 2000s required fast implementations on <a href="/wiki/Graphics_processing_unit" title="Graphics processing unit">graphics processing units</a> (GPUs). In 2004, it was shown by K. S. Oh and K. Jung that standard neural networks can be greatly accelerated on GPUs. Their implementation was 20 times faster than an equivalent implementation on <a href="/wiki/CPU" class="mw-redirect" title="CPU">CPU</a>.<a href="#cite_note-58">[58]</a> In 2005, another paper also emphasised the value of <a href="/wiki/GPGPU" class="mw-redirect" title="GPGPU">GPGPU</a> for <a href="/wiki/Machine_learning" title="Machine learning">machine learning</a>.<a href="#cite_note-59">[59]</a> The first GPU-implementation of a CNN was described in 2006 by K. Chellapilla et al. Their implementation was 4 times faster than an equivalent implementation on CPU.<a href="#cite_note-60">[60]</a> In the same period, GPUs were also used for unsupervised training of <a href="/wiki/Deep_belief_network" title="Deep belief network">deep belief networks</a>.<a href="#cite_note-61">[61]</a><a href="#cite_note-62">[62]</a><a href="#cite_note-63">[63]</a><a href="#cite_note-LSD_1-64">[64]</a> In 2010, Dan Ciresan et al. at <a href="/wiki/IDSIA" class="mw-redirect" title="IDSIA">IDSIA</a> trained deep feedforward networks on GPUs.<a href="#cite_note-65">[65]</a> In 2011, they extended this to CNNs, accelerating by 60 compared to training CPU.<a href="#cite_note-flexible-24">[24]</a> In 2011, the network win an image recognition contest where they achieved superhuman performance for the first time.<a href="#cite_note-66">[66]</a> Then they won more competitions and achieved state of the art on several benchmarks.<a href="#cite_note-67">[67]</a><a href="#cite_note-schdeepscholar-42">[42]</a><a href="#cite_note-mcdns-27">[27]</a> Subsequently, <a href="/wiki/AlexNet" title="AlexNet">AlexNet</a>, a similar GPU-based CNN by Alex Krizhevsky et al. won the <a href="/wiki/ImageNet_Large_Scale_Visual_Recognition_Challenge" class="mw-redirect" title="ImageNet Large Scale Visual Recognition Challenge">ImageNet Large Scale Visual Recognition Challenge</a> 2012.<a href="#cite_note-:02-68">[68]</a> It was an early catalytic event for the <a href="/wiki/AI_boom" title="AI boom">AI boom</a>. Compared to the training of CNNs using <a href="/wiki/GPU" class="mw-redirect" title="GPU">GPUs</a>, not much attention was given to CPU. (Viebke et al 2019) parallelizes CNN by thread- and <a href="/wiki/SIMD" class="mw-redirect" title="SIMD">SIMD</a>-level parallelism that is available on the <a href="/wiki/Xeon_Phi" title="Xeon Phi">Intel Xeon Phi</a>.<a href="#cite_note-69">[69]</a><a href="#cite_note-70">[70]</a> <div class="mw-heading mw-heading2"><h2 id="Distinguishing_features">Distinguishing features</h2>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=18" title="Edit section: Distinguishing features">edit</a>]</div> In the past, traditional <a href="/wiki/Multilayer_perceptron" title="Multilayer perceptron">multilayer perceptron</a> (MLP) models were used for image recognition.[<a href="/wiki/Wikipedia:AUDIENCE" class="mw-redirect" title="Wikipedia:AUDIENCE">example needed</a>] However, the full connectivity between nodes caused the <a href="/wiki/Curse_of_dimensionality" title="Curse of dimensionality">curse of dimensionality</a>, and was computationally intractable with higher-resolution images. A 1000×1000-pixel image with <a href="/wiki/RGB_color_model" title="RGB color model">RGB color</a> channels has 3 million weights per fully-connected neuron, which is too high to feasibly process efficiently at scale. <figure class="mw-halign-left" typeof="mw:File/Thumb"><a href="/wiki/File:Conv_layers.png" class="mw-file-description"><img src="//upload.wikimedia.org/wikipedia/commons/thumb/8/8a/Conv_layers.png/237px-Conv_layers.png" decoding="async" width="237" height="130" class="mw-file-element" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/8/8a/Conv_layers.png/356px-Conv_layers.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/8/8a/Conv_layers.png/474px-Conv_layers.png 2x" data-file-width="567" data-file-height="310" /></a><figcaption>CNN layers arranged in 3 dimensions</figcaption></figure> For example, in <a href="/wiki/CIFAR-10" title="CIFAR-10">CIFAR-10</a>, images are only of size 32×32×3 (32 wide, 32 high, 3 color channels), so a single fully connected neuron in the first hidden layer of a regular neural network would have 32*32*3 = 3,072 weights. A 200×200 image, however, would lead to neurons that have 200*200*3 = 120,000 weights. Also, such network architecture does not take into account the spatial structure of data, treating input pixels which are far apart in the same way as pixels that are close together. This ignores <a href="/wiki/Locality_of_reference" title="Locality of reference">locality of reference</a> in data with a grid-topology (such as images), both computationally and semantically. Thus, full connectivity of neurons is wasteful for purposes such as image recognition that are dominated by <a href="/wiki/Spatial_locality" class="mw-redirect" title="Spatial locality">spatially local</a> input patterns. Convolutional neural networks are variants of multilayer perceptrons, designed to emulate the behavior of a <a href="/wiki/Visual_cortex" title="Visual cortex">visual cortex</a>. These models mitigate the challenges posed by the MLP architecture by exploiting the strong spatially local correlation present in natural images. As opposed to MLPs, CNNs have the following distinguishing features: <ul><li>3D volumes of neurons. The layers of a CNN have neurons arranged in <a href="/wiki/Three-dimensional_space" title="Three-dimensional space">3 dimensions</a>: width, height and depth.<a href="#cite_note-71">[71]</a> Where each neuron inside a convolutional layer is connected to only a small region of the layer before it, called a receptive field. Distinct types of layers, both locally and completely connected, are stacked to form a CNN architecture.</li> <li>Local connectivity: following the concept of receptive fields, CNNs exploit spatial locality by enforcing a local connectivity pattern between neurons of adjacent layers. The architecture thus ensures that the learned "<a href="/wiki/Filter_(signal_processing)" title="Filter (signal processing)">filters</a>" produce the strongest response to a spatially local input pattern. Stacking many such layers leads to <a href="/wiki/Nonlinear_filter" title="Nonlinear filter">nonlinear filters</a> that become increasingly global (i.e. responsive to a larger region of pixel space) so that the network first creates representations of small parts of the input, then from them assembles representations of larger areas.</li> <li>Shared weights: In CNNs, each filter is replicated across the entire visual field. These replicated units share the same parameterization (weight vector and bias) and form a feature map. This means that all the neurons in a given convolutional layer respond to the same feature within their specific response field. Replicating units in this way allows for the resulting activation map to be <a href="/wiki/Equivariant_map" title="Equivariant map">equivariant</a> under shifts of the locations of input features in the visual field, i.e. they grant translational <a href="/wiki/Equivariant_map" title="Equivariant map">equivariance</a>—given that the layer has a stride of one.<a href="#cite_note-:5-72">[72]</a></li> <li>Pooling: In a CNN's <a href="/wiki/Pooling_layer" title="Pooling layer">pooling layers</a>, feature maps are divided into rectangular sub-regions, and the features in each rectangle are independently down-sampled to a single value, commonly by taking their average or maximum value. In addition to reducing the sizes of feature maps, the pooling operation grants a degree of local <a href="/wiki/Translational_symmetry" title="Translational symmetry">translational invariance</a> to the features contained therein, allowing the CNN to be more robust to variations in their positions.<a href="#cite_note-:6-15">[15]</a></li></ul> Together, these properties allow CNNs to achieve better generalization on <a href="/wiki/Computer_vision" title="Computer vision">vision problems</a>. Weight sharing dramatically reduces the number of <a href="/wiki/Free_parameter" title="Free parameter">free parameters</a> learned, thus lowering the memory requirements for running the network and allowing the training of larger, more powerful networks. <div class="mw-heading mw-heading2"><h2 id="Building_blocks">Building blocks</h2>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=19" title="Edit section: Building blocks">edit</a>]</div> A CNN architecture is formed by a stack of distinct layers that transform the input volume into an output volume (e.g. holding the class scores) through a differentiable function. A few distinct types of layers are commonly used. These are further discussed below.<figure class="mw-halign-left" typeof="mw:File/Thumb"><a href="/wiki/File:Conv_layer.png" class="mw-file-description"><img src="//upload.wikimedia.org/wikipedia/commons/thumb/6/68/Conv_layer.png/229px-Conv_layer.png" decoding="async" width="229" height="154" class="mw-file-element" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/6/68/Conv_layer.png/344px-Conv_layer.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/6/68/Conv_layer.png/458px-Conv_layer.png 2x" data-file-width="634" data-file-height="426" /></a><figcaption>Neurons of a convolutional layer (blue), connected to their receptive field (red)</figcaption></figure> <div class="mw-heading mw-heading3"><h3 id="Convolutional_layer">Convolutional layer</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=20" title="Edit section: Convolutional layer">edit</a>]</div> <figure class="mw-default-size" typeof="mw:File/Thumb"><a href="/wiki/File:Convolutional_neural_network,_convolution_worked_example.png" class="mw-file-description"><img src="//upload.wikimedia.org/wikipedia/commons/thumb/b/bd/Convolutional_neural_network%2C_convolution_worked_example.png/220px-Convolutional_neural_network%2C_convolution_worked_example.png" decoding="async" width="220" height="154" class="mw-file-element" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/b/bd/Convolutional_neural_network%2C_convolution_worked_example.png/330px-Convolutional_neural_network%2C_convolution_worked_example.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/b/bd/Convolutional_neural_network%2C_convolution_worked_example.png/440px-Convolutional_neural_network%2C_convolution_worked_example.png 2x" data-file-width="1426" data-file-height="1000" /></a><figcaption>A worked example of performing a convolution. The convolution has stride 1, zero-padding, with kernel size 3-by-3. The convolution kernel is a <a href="/wiki/Discrete_Laplace_operator" title="Discrete Laplace operator">discrete Laplacian operator</a>.</figcaption></figure> The convolutional layer is the core building block of a CNN. The layer's parameters consist of a set of learnable <a href="/wiki/Filter_(signal_processing)" title="Filter (signal processing)">filters</a> (or <a href="/wiki/Kernel_(image_processing)" title="Kernel (image processing)">kernels</a>), which have a small receptive field, but extend through the full depth of the input volume. During the forward pass, each filter is <a href="/wiki/Convolution" title="Convolution">convolved</a> across the width and height of the input volume, computing the <a href="/wiki/Dot_product" title="Dot product">dot product</a> between the filter entries and the input, producing a 2-dimensional <a href="/wiki/Activation_function" title="Activation function">activation map</a> of that filter. As a result, the network learns filters that activate when it detects some specific type of <a href="/wiki/Feature_(machine_learning)" title="Feature (machine learning)">feature</a> at some spatial position in the input.<a href="#cite_note-Géron_Hands-on_ML_2019-73">[73]</a><a href="#cite_note-74">[nb 1]</a> Stacking the activation maps for all filters along the depth dimension forms the full output volume of the convolution layer. Every entry in the output volume can thus also be interpreted as an output of a neuron that looks at a small region in the input. Each entry in an activation map use the same set of parameters that define the filter. <a href="/wiki/Self-supervised_learning" title="Self-supervised learning">Self-supervised learning</a> has been adapted for use in convolutional layers by using sparse patches with a high-mask ratio and a global response normalization layer.[<a href="/wiki/Wikipedia:Citation_needed" title="Wikipedia:Citation needed">citation needed</a>] <div class="mw-heading mw-heading4"><h4 id="Local_connectivity">Local connectivity</h4>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=21" title="Edit section: Local connectivity">edit</a>]</div> <figure typeof="mw:File/Thumb"><a href="/wiki/File:Typical_cnn.png" class="mw-file-description"><img src="//upload.wikimedia.org/wikipedia/commons/thumb/6/63/Typical_cnn.png/395px-Typical_cnn.png" decoding="async" width="395" height="122" class="mw-file-element" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/6/63/Typical_cnn.png/593px-Typical_cnn.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/6/63/Typical_cnn.png/790px-Typical_cnn.png 2x" data-file-width="1040" data-file-height="320" /></a><figcaption>Typical CNN architecture</figcaption></figure> When dealing with high-dimensional inputs such as images, it is impractical to connect neurons to all neurons in the previous volume because such a network architecture does not take the spatial structure of the data into account. Convolutional networks exploit spatially local correlation by enforcing a <a href="/wiki/Sparse_network" title="Sparse network">sparse local connectivity</a> pattern between neurons of adjacent layers: each neuron is connected to only a small region of the input volume. The extent of this connectivity is a <a href="/wiki/Hyperparameter_optimization" title="Hyperparameter optimization">hyperparameter</a> called the <a href="/wiki/Receptive_field" title="Receptive field">receptive field</a> of the neuron. The connections are <a href="/wiki/Spatial_locality" class="mw-redirect" title="Spatial locality">local in space</a> (along width and height), but always extend along the entire depth of the input volume. Such an architecture ensures that the learned (<a href="/wiki/British_English_language" class="mw-redirect" title="British English language">British English</a>: learnt) filters produce the strongest response to a spatially local input pattern. <div class="mw-heading mw-heading4"><h4 id="Spatial_arrangement">Spatial arrangement</h4>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=22" title="Edit section: Spatial arrangement">edit</a>]</div> Three <a href="/wiki/Hyperparameter_(machine_learning)" title="Hyperparameter (machine learning)">hyperparameters</a> control the size of the output volume of the convolutional layer: the depth, <a href="/wiki/Stride_of_an_array" title="Stride of an array">stride</a>, and padding size: <ul><li>The depth of the output volume controls the number of neurons in a layer that connect to the same region of the input volume. These neurons learn to activate for different features in the input. For example, if the first convolutional layer takes the raw image as input, then different neurons along the depth dimension may activate in the presence of various oriented edges, or blobs of color.</li> <li>Stride controls how depth columns around the width and height are allocated. If the stride is 1, then we move the filters one pixel at a time. This leads to heavily <a href="/wiki/Intersection_(set_theory)" title="Intersection (set theory)">overlapping</a> receptive fields between the columns, and to large output volumes. For any integer <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\textstyle S>0,}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="false" scriptlevel="0"> <mi>S</mi> <mo>></mo> <mn>0</mn> <mo>,</mo> </mstyle> </mrow> <annotation encoding="application/x-tex">{\textstyle S>0,}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/d1fb1c59de56b22cfdf94e6c2a03f4b0dfc7c4d2" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.671ex; width:6.407ex; height:2.509ex;" alt="{\textstyle S>0,}"> a stride S means that the filter is translated S units at a time per output. In practice, <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\textstyle S\geq 3}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="false" scriptlevel="0"> <mi>S</mi> <mo>≥</mo> <mn>3</mn> </mstyle> </mrow> <annotation encoding="application/x-tex">{\textstyle S\geq 3}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/1b1cc974b8af65e3a94aa730c0f360dd7059604d" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.505ex; width:5.76ex; height:2.343ex;" alt="{\textstyle S\geq 3}"> is rare. A greater stride means smaller overlap of receptive fields and smaller spatial dimensions of the output volume.<a href="#cite_note-75">[74]</a></li> <li>Sometimes, it is convenient to pad the input with zeros (or other values, such as the average of the region) on the border of the input volume. The size of this padding is a third hyperparameter. Padding provides control of the output volume's spatial size. In particular, sometimes it is desirable to exactly preserve the spatial size of the input volume, this is commonly referred to as "same" padding.</li></ul> <figure typeof="mw:File/Thumb"><a href="/wiki/File:Convolutional_neural_network,_boundary_conditions.png" class="mw-file-description"><img src="//upload.wikimedia.org/wikipedia/commons/thumb/7/7f/Convolutional_neural_network%2C_boundary_conditions.png/315px-Convolutional_neural_network%2C_boundary_conditions.png" decoding="async" width="315" height="105" class="mw-file-element" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/7/7f/Convolutional_neural_network%2C_boundary_conditions.png/473px-Convolutional_neural_network%2C_boundary_conditions.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/7/7f/Convolutional_neural_network%2C_boundary_conditions.png/630px-Convolutional_neural_network%2C_boundary_conditions.png 2x" data-file-width="1426" data-file-height="475" /></a><figcaption>Three example padding conditions. Replication condition means that the pixel outside is padded with the closest pixel inside. The reflection padding is where the pixel outside is padded with the pixel inside, reflected across the boundary of the image. The circular padding is where the pixel outside wraps around to the other side of the image.</figcaption></figure> The spatial size of the output volume is a function of the input volume size <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle W}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mi>W</mi> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle W}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/54a9c4c547f4d6111f81946cad242b18298d70b7" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.338ex; width:2.435ex; height:2.176ex;" alt="{\displaystyle W}">, the kernel field size <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle K}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mi>K</mi> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle K}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/2b76fce82a62ed5461908f0dc8f037de4e3686b0" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.338ex; width:2.066ex; height:2.176ex;" alt="{\displaystyle K}"> of the convolutional layer neurons, the stride <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle S}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mi>S</mi> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle S}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/4611d85173cd3b508e67077d4a1252c9c05abca2" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.338ex; width:1.499ex; height:2.176ex;" alt="{\displaystyle S}">, and the amount of zero padding <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle P}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mi>P</mi> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle P}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/b4dc73bf40314945ff376bd363916a738548d40a" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.338ex; width:1.745ex; height:2.176ex;" alt="{\displaystyle P}"> on the border. The number of neurons that "fit" in a given volume is then: <dl><dd><math display="block" xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle {\frac {W-K+2P}{S}}+1.}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mrow class="MJX-TeXAtom-ORD"> <mfrac> <mrow> <mi>W</mi> <mo>−</mo> <mi>K</mi> <mo>+</mo> <mn>2</mn> <mi>P</mi> </mrow> <mi>S</mi> </mfrac> </mrow> <mo>+</mo> <mn>1.</mn> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle {\frac {W-K+2P}{S}}+1.}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/a5325dc0b1b8695f19ec8fe3485d0da19040c622" class="mwe-math-fallback-image-display mw-invert skin-invert" aria-hidden="true" style="vertical-align: -2.005ex; width:18.576ex; height:5.343ex;" alt="{\displaystyle {\frac {W-K+2P}{S}}+1.}"></dd></dl> If this number is not an <a href="/wiki/Integer" title="Integer">integer</a>, then the strides are incorrect and the neurons cannot be tiled to fit across the input volume in a <a href="/wiki/Symmetry" title="Symmetry">symmetric</a> way. In general, setting zero padding to be <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\textstyle P=(K-1)/2}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="false" scriptlevel="0"> <mi>P</mi> <mo>=</mo> <mo stretchy="false">(</mo> <mi>K</mi> <mo>−</mo> <mn>1</mn> <mo stretchy="false">)</mo> <mrow class="MJX-TeXAtom-ORD"> <mo>/</mo> </mrow> <mn>2</mn> </mstyle> </mrow> <annotation encoding="application/x-tex">{\textstyle P=(K-1)/2}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/4b1c80a44e8d2e568cb4adfc6f6c8bf554f210e6" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.838ex; width:15.047ex; height:2.843ex;" alt="{\textstyle P=(K-1)/2}"> when the stride is <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle S=1}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mi>S</mi> <mo>=</mo> <mn>1</mn> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle S=1}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/d5e2b58c1aaaf2718fb801e97bf21d1f72726372" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.338ex; width:5.76ex; height:2.176ex;" alt="{\displaystyle S=1}"> ensures that the input volume and output volume will have the same size spatially. However, it is not always completely necessary to use all of the neurons of the previous layer. For example, a neural network designer may decide to use just a portion of padding. <div class="mw-heading mw-heading4"><h4 id="Parameter_sharing">Parameter sharing</h4>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=23" title="Edit section: Parameter sharing">edit</a>]</div> A parameter sharing scheme is used in convolutional layers to control the number of free parameters. It relies on the assumption that if a patch feature is useful to compute at some spatial position, then it should also be useful to compute at other positions. Denoting a single 2-dimensional slice of depth as a depth slice, the neurons in each depth slice are constrained to use the same weights and bias. Since all neurons in a single depth slice share the same parameters, the forward pass in each depth slice of the convolutional layer can be computed as a <a href="/wiki/Convolution" title="Convolution">convolution</a> of the neuron's weights with the input volume.<a href="#cite_note-76">[nb 2]</a> Therefore, it is common to refer to the sets of weights as a filter (or a <a href="/wiki/Kernel_(image_processing)" title="Kernel (image processing)">kernel</a>), which is convolved with the input. The result of this convolution is an <a href="/wiki/Activation_function" title="Activation function">activation map</a>, and the set of activation maps for each different filter are stacked together along the depth dimension to produce the output volume. Parameter sharing contributes to the <a href="/wiki/Translational_symmetry" title="Translational symmetry">translation invariance</a> of the CNN architecture.<a href="#cite_note-:6-15">[15]</a> Sometimes, the parameter sharing assumption may not make sense. This is especially the case when the input images to a CNN have some specific centered structure; for which we expect completely different features to be learned on different spatial locations. One practical example is when the inputs are faces that have been centered in the image: we might expect different eye-specific or hair-specific features to be learned in different parts of the image. In that case it is common to relax the parameter sharing scheme, and instead simply call the layer a "locally connected layer". <div class="mw-heading mw-heading3"><h3 id="Pooling_layer">Pooling layer</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=24" title="Edit section: Pooling layer">edit</a>]</div> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1236090951"><div role="note" class="hatnote navigation-not-searchable">Main article: <a href="/wiki/Pooling_layer" title="Pooling layer">Pooling layer</a></div> <figure typeof="mw:File/Thumb"><a href="/wiki/File:Convolutional_neural_network,_maxpooling.png" class="mw-file-description"><img src="//upload.wikimedia.org/wikipedia/commons/thumb/b/bb/Convolutional_neural_network%2C_maxpooling.png/311px-Convolutional_neural_network%2C_maxpooling.png" decoding="async" width="311" height="120" class="mw-file-element" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/b/bb/Convolutional_neural_network%2C_maxpooling.png/467px-Convolutional_neural_network%2C_maxpooling.png 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/b/bb/Convolutional_neural_network%2C_maxpooling.png/622px-Convolutional_neural_network%2C_maxpooling.png 2x" data-file-width="1426" data-file-height="551" /></a><figcaption>Worked example of 2x2 maxpooling with stride 2.</figcaption></figure> <figure typeof="mw:File/Thumb"><a href="/wiki/File:Max_pooling.png" class="mw-file-description"><img src="//upload.wikimedia.org/wikipedia/commons/thumb/e/e9/Max_pooling.png/314px-Max_pooling.png" decoding="async" width="314" height="182" class="mw-file-element" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/e/e9/Max_pooling.png/471px-Max_pooling.png 1.5x, //upload.wikimedia.org/wikipedia/commons/e/e9/Max_pooling.png 2x" data-file-width="570" data-file-height="330" /></a><figcaption>Max pooling with a 2x2 filter and stride = 2</figcaption></figure> Another important concept of CNNs is pooling, which is used as a form of non-linear <a href="/wiki/Downsampling_(signal_processing)" title="Downsampling (signal processing)">down-sampling</a>. Pooling provides downsampling because it reduces the spatial dimensions (height and width) of the input feature maps while retaining the most important information. There are several non-linear functions to implement pooling, where max pooling and average pooling are the most common. Pooling aggregates information from small regions of the input creating <a href="/wiki/Partition_of_a_set" title="Partition of a set">partitions</a> of the input feature map, typically using a fixed-size window (like 2x2) and applying a stride (often 2) to move the window across the input.<a href="#cite_note-77">[75]</a> Note that without using a stride greater than 1, pooling would not perform downsampling, as it would simply move the pooling window across the input one step at a time, without reducing the size of the feature map. In other words, the stride is what actually causes the downsampling by determining how much the pooling window moves over the input. Intuitively, the exact location of a feature is less important than its rough location relative to other features. This is the idea behind the use of pooling in convolutional neural networks. The pooling layer serves to progressively reduce the spatial size of the representation, to reduce the number of parameters, <a href="/wiki/Memory_footprint" title="Memory footprint">memory footprint</a> and amount of computation in the network, and hence to also control <a href="/wiki/Overfitting" title="Overfitting">overfitting</a>. This is known as down-sampling. It is common to periodically insert a pooling layer between successive convolutional layers (each one typically followed by an activation function, such as a <a href="#ReLU_layer">ReLU layer</a>) in a CNN architecture.<a href="#cite_note-Géron_Hands-on_ML_2019-73">[73]</a>: 460–461  While pooling layers contribute to local translation invariance, they do not provide global translation invariance in a CNN, unless a form of global pooling is used.<a href="#cite_note-:6-15">[15]</a><a href="#cite_note-:5-72">[72]</a> The pooling layer commonly operates independently on every depth, or slice, of the input and resizes it spatially. A very common form of max pooling is a layer with filters of size 2×2, applied with a stride of 2, which subsamples every depth slice in the input by 2 along both width and height, discarding 75% of the activations:<math display="block" xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle f_{X,Y}(S)=\max _{a,b=0}^{1}S_{2X+a,2Y+b}.}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <msub> <mi>f</mi> <mrow class="MJX-TeXAtom-ORD"> <mi>X</mi> <mo>,</mo> <mi>Y</mi> </mrow> </msub> <mo stretchy="false">(</mo> <mi>S</mi> <mo stretchy="false">)</mo> <mo>=</mo> <munderover> <mo movablelimits="true" form="prefix">max</mo> <mrow class="MJX-TeXAtom-ORD"> <mi>a</mi> <mo>,</mo> <mi>b</mi> <mo>=</mo> <mn>0</mn> </mrow> <mrow class="MJX-TeXAtom-ORD"> <mn>1</mn> </mrow> </munderover> <msub> <mi>S</mi> <mrow class="MJX-TeXAtom-ORD"> <mn>2</mn> <mi>X</mi> <mo>+</mo> <mi>a</mi> <mo>,</mo> <mn>2</mn> <mi>Y</mi> <mo>+</mo> <mi>b</mi> </mrow> </msub> <mo>.</mo> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle f_{X,Y}(S)=\max _{a,b=0}^{1}S_{2X+a,2Y+b}.}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/fb35d034c37054cf3920f0dc52236a636dd8f902" class="mwe-math-fallback-image-display mw-invert skin-invert" aria-hidden="true" style="vertical-align: -2.338ex; width:26.794ex; height:5.509ex;" alt="{\displaystyle f_{X,Y}(S)=\max _{a,b=0}^{1}S_{2X+a,2Y+b}.}"> In this case, every <a href="/wiki/Maximum" class="mw-redirect" title="Maximum">max operation</a> is over 4 numbers. The depth dimension remains unchanged (this is true for other forms of pooling as well). In addition to max pooling, pooling units can use other functions, such as <a href="/wiki/Average" title="Average">average</a> pooling or <a href="/wiki/Euclidean_norm" class="mw-redirect" title="Euclidean norm">ℓ2-norm</a> pooling. Average pooling was often used historically but has recently fallen out of favor compared to max pooling, which generally performs better in practice.<a href="#cite_note-Scherer-ICANN-2010-78">[76]</a> Due to the effects of fast spatial reduction of the size of the representation,[<a href="/wiki/Wikipedia:Avoid_weasel_words" class="mw-redirect" title="Wikipedia:Avoid weasel words">which?</a>] there is a recent trend towards using smaller filters<a href="#cite_note-79">[77]</a> or discarding pooling layers altogether.<a href="#cite_note-80">[78]</a> <figure typeof="mw:File/Thumb"><a href="/wiki/File:RoI_pooling_animated.gif" class="mw-file-description"><img src="//upload.wikimedia.org/wikipedia/commons/thumb/d/dc/RoI_pooling_animated.gif/400px-RoI_pooling_animated.gif" decoding="async" width="400" height="300" class="mw-file-element" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/d/dc/RoI_pooling_animated.gif/600px-RoI_pooling_animated.gif 1.5x, //upload.wikimedia.org/wikipedia/commons/d/dc/RoI_pooling_animated.gif 2x" data-file-width="800" data-file-height="600" /></a><figcaption>RoI pooling to size 2x2. In this example region proposal (an input parameter) has size 7x5.</figcaption></figure> <div class="mw-heading mw-heading4"><h4 id="Channel_max_pooling">Channel max pooling</h4>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=25" title="Edit section: Channel max pooling">edit</a>]</div> A channel max pooling (CMP) operation layer conducts the MP operation along the channel side among the corresponding positions of the consecutive feature maps for the purpose of redundant information elimination. The CMP makes the significant features gather together within fewer channels, which is important for fine-grained image classification that needs more discriminating features. Meanwhile, another advantage of the CMP operation is to make the channel number of feature maps smaller before it connects to the first fully connected (FC) layer. Similar to the MP operation, we denote the input feature maps and output feature maps of a CMP layer as F ∈ R(C×M×N) and C ∈ R(c×M×N), respectively, where C and c are the channel numbers of the input and output feature maps, M and N are the widths and the height of the feature maps, respectively. Note that the CMP operation only changes the channel number of the feature maps. The width and the height of the feature maps are not changed, which is different from the MP operation.<a href="#cite_note-Ma_Chang_Xie_Ding_2019_pp._3224–3233-81">[79]</a> See <a href="#cite_note-82">[80]</a><a href="#cite_note-83">[81]</a> for reviews for pooling methods. <div class="mw-heading mw-heading3"><h3 id="ReLU_layer">ReLU layer</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=26" title="Edit section: ReLU layer">edit</a>]</div> ReLU is the abbreviation of <a href="/wiki/Rectifier_(neural_networks)" title="Rectifier (neural networks)">rectified linear unit</a>. It was proposed by <a href="/wiki/Alston_Scott_Householder" title="Alston Scott Householder">Alston Householder</a> in 1941,<a href="#cite_note-84">[82]</a> and used in CNN by <a href="/wiki/Kunihiko_Fukushima" title="Kunihiko Fukushima">Kunihiko Fukushima</a> in 1969.<a href="#cite_note-Fukushima1969-38">[38]</a> ReLU applies the non-saturating <a href="/wiki/Activation_function" title="Activation function">activation function</a> <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\textstyle f(x)=\max(0,x)}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="false" scriptlevel="0"> <mi>f</mi> <mo stretchy="false">(</mo> <mi>x</mi> <mo stretchy="false">)</mo> <mo>=</mo> <mo movablelimits="true" form="prefix">max</mo> <mo stretchy="false">(</mo> <mn>0</mn> <mo>,</mo> <mi>x</mi> <mo stretchy="false">)</mo> </mstyle> </mrow> <annotation encoding="application/x-tex">{\textstyle f(x)=\max(0,x)}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/e2559111e4c3500df55d16f958f853e19e90804a" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.838ex; width:17.177ex; height:2.843ex;" alt="{\textstyle f(x)=\max(0,x)}">.<a href="#cite_note-:02-68">[68]</a> It effectively removes negative values from an activation map by setting them to zero.<a href="#cite_note-Romanuke4-85">[83]</a> It introduces <a href="/wiki/Nonlinearity_(disambiguation)" class="mw-disambig" title="Nonlinearity (disambiguation)">nonlinearity</a> to the <a href="/wiki/Decision_boundary" title="Decision boundary">decision function</a> and in the overall network without affecting the receptive fields of the convolution layers. In 2011, Xavier Glorot, Antoine Bordes and <a href="/wiki/Yoshua_Bengio" title="Yoshua Bengio">Yoshua Bengio</a> found that ReLU enables better training of deeper networks,<a href="#cite_note-glorot2011-86">[84]</a> compared to widely used activation functions prior to 2011. Other functions can also be used to increase nonlinearity, for example the saturating <a href="/wiki/Hyperbolic_tangent" class="mw-redirect" title="Hyperbolic tangent">hyperbolic tangent</a> <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle f(x)=\tanh(x)}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mi>f</mi> <mo stretchy="false">(</mo> <mi>x</mi> <mo stretchy="false">)</mo> <mo>=</mo> <mi>tanh</mi> <mo>⁡</mo> <mo stretchy="false">(</mo> <mi>x</mi> <mo stretchy="false">)</mo> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle f(x)=\tanh(x)}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/1a319ec32dbb0c625fa4802baf9252d1f00854e2" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.838ex; width:15.307ex; height:2.843ex;" alt="{\displaystyle f(x)=\tanh(x)}">, <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle f(x)=|\tanh(x)|}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mi>f</mi> <mo stretchy="false">(</mo> <mi>x</mi> <mo stretchy="false">)</mo> <mo>=</mo> <mrow class="MJX-TeXAtom-ORD"> <mo stretchy="false">|</mo> </mrow> <mi>tanh</mi> <mo>⁡</mo> <mo stretchy="false">(</mo> <mi>x</mi> <mo stretchy="false">)</mo> <mrow class="MJX-TeXAtom-ORD"> <mo stretchy="false">|</mo> </mrow> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle f(x)=|\tanh(x)|}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/d1eb71e39ce9687851b7ec55bb8f54f42df2a828" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.838ex; width:16.988ex; height:2.843ex;" alt="{\displaystyle f(x)=|\tanh(x)|}">, and the <a href="/wiki/Sigmoid_function" title="Sigmoid function">sigmoid function</a> <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\textstyle \sigma (x)=(1+e^{-x})^{-1}}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="false" scriptlevel="0"> <mi>σ</mi> <mo stretchy="false">(</mo> <mi>x</mi> <mo stretchy="false">)</mo> <mo>=</mo> <mo stretchy="false">(</mo> <mn>1</mn> <mo>+</mo> <msup> <mi>e</mi> <mrow class="MJX-TeXAtom-ORD"> <mo>−</mo> <mi>x</mi> </mrow> </msup> <msup> <mo stretchy="false">)</mo> <mrow class="MJX-TeXAtom-ORD"> <mo>−</mo> <mn>1</mn> </mrow> </msup> </mstyle> </mrow> <annotation encoding="application/x-tex">{\textstyle \sigma (x)=(1+e^{-x})^{-1}}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/3805da830a1a77523aefcdae0d2164d954849066" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.838ex; width:19.247ex; height:3.009ex;" alt="{\textstyle \sigma (x)=(1+e^{-x})^{-1}}">. ReLU is often preferred to other functions because it trains the neural network several times faster without a significant penalty to <a href="/wiki/Generalization_(learning)" title="Generalization (learning)">generalization</a> accuracy.<a href="#cite_note-87">[85]</a> <div class="mw-heading mw-heading3"><h3 id="Fully_connected_layer">Fully connected layer</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=27" title="Edit section: Fully connected layer">edit</a>]</div> After several convolutional and max pooling layers, the final classification is done via fully connected layers. Neurons in a fully connected layer have connections to all activations in the previous layer, as seen in regular (non-convolutional) <a href="/wiki/Artificial_neural_network" class="mw-redirect" title="Artificial neural network">artificial neural networks</a>. Their activations can thus be computed as an <a href="/wiki/Affine_transformation" title="Affine transformation">affine transformation</a>, with <a href="/wiki/Matrix_multiplication" title="Matrix multiplication">matrix multiplication</a> followed by a bias offset (<a href="/wiki/Vector_addition" class="mw-redirect" title="Vector addition">vector addition</a> of a learned or fixed bias term). <div class="mw-heading mw-heading3"><h3 id="Loss_layer">Loss layer</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=28" title="Edit section: Loss layer">edit</a>]</div> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1236090951"><div role="note" class="hatnote navigation-not-searchable">Main articles: <a href="/wiki/Loss_function" title="Loss function">Loss function</a> and <a href="/wiki/Loss_functions_for_classification" title="Loss functions for classification">Loss functions for classification</a></div> The "loss layer", or "<a href="/wiki/Loss_function" title="Loss function">loss function</a>", specifies how <a href="/wiki/Training" title="Training">training</a> penalizes the deviation between the predicted output of the network, and the <a href="/wiki/Ground_truth" title="Ground truth">true</a> data labels (during supervised learning). Various <a href="/wiki/Loss_function" title="Loss function">loss functions</a> can be used, depending on the specific task. The <a href="/wiki/Softmax_function" title="Softmax function">Softmax</a> loss function is used for predicting a single class of K mutually exclusive classes.<a href="#cite_note-88">[nb 3]</a> <a href="/wiki/Sigmoid_function" title="Sigmoid function">Sigmoid</a> <a href="/wiki/Cross_entropy" class="mw-redirect" title="Cross entropy">cross-entropy</a> loss is used for predicting K independent probability values in <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle [0,1]}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mo stretchy="false">[</mo> <mn>0</mn> <mo>,</mo> <mn>1</mn> <mo stretchy="false">]</mo> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle [0,1]}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/738f7d23bb2d9642bab520020873cccbef49768d" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.838ex; width:4.653ex; height:2.843ex;" alt="{\displaystyle [0,1]}">. <a href="/wiki/Euclidean_distance" title="Euclidean distance">Euclidean</a> loss is used for <a href="/wiki/Regression_(machine_learning)" class="mw-redirect" title="Regression (machine learning)">regressing</a> to <a href="/wiki/Real_number" title="Real number">real-valued</a> labels <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle (-\infty ,\infty )}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mo stretchy="false">(</mo> <mo>−</mo> <mi mathvariant="normal">∞</mi> <mo>,</mo> <mi mathvariant="normal">∞</mi> <mo stretchy="false">)</mo> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle (-\infty ,\infty )}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/0c8c11c44279888c9e395eeb5f45d121348ae10a" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.838ex; width:9.299ex; height:2.843ex;" alt="{\displaystyle (-\infty ,\infty )}">. <div class="mw-heading mw-heading2"><h2 id="Hyperparameters">Hyperparameters</h2>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=29" title="Edit section: Hyperparameters">edit</a>]</div> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1251242444"><table class="box-More_citations_needed_section plainlinks metadata ambox ambox-content ambox-Refimprove" role="presentation"><tbody><tr><td class="mbox-image"><div class="mbox-image-div"><a href="/wiki/File:Question_book-new.svg" class="mw-file-description"><img alt="" src="//upload.wikimedia.org/wikipedia/en/thumb/9/99/Question_book-new.svg/50px-Question_book-new.svg.png" decoding="async" width="50" height="39" class="mw-file-element" srcset="//upload.wikimedia.org/wikipedia/en/thumb/9/99/Question_book-new.svg/75px-Question_book-new.svg.png 1.5x, //upload.wikimedia.org/wikipedia/en/thumb/9/99/Question_book-new.svg/100px-Question_book-new.svg.png 2x" data-file-width="512" data-file-height="399" /></a></div></td><td class="mbox-text"><div class="mbox-text-span">This section needs additional citations for <a href="/wiki/Wikipedia:Verifiability" title="Wikipedia:Verifiability">verification</a>. Please help <a href="/wiki/Special:EditPage/Convolutional_neural_network" title="Special:EditPage/Convolutional neural network">improve this article</a> by <a href="/wiki/Help:Referencing_for_beginners" title="Help:Referencing for beginners">adding citations to reliable sources</a> in this section. Unsourced material may be challenged and removed. (June 2017) (<a href="/wiki/Help:Maintenance_template_removal" title="Help:Maintenance template removal">Learn how and when to remove this message</a>)</div></td></tr></tbody></table> Hyperparameters are various settings that are used to control the learning process. CNNs use more <a href="/wiki/Hyperparameter_(machine_learning)" title="Hyperparameter (machine learning)">hyperparameters</a> than a standard multilayer perceptron (MLP). <div class="mw-heading mw-heading3"><h3 id="Kernel_size">Kernel size</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=30" title="Edit section: Kernel size">edit</a>]</div> The kernel is the number of pixels processed together. It is typically expressed as the kernel's dimensions, e.g., 2x2, or 3x3. <div class="mw-heading mw-heading3"><h3 id="Padding">Padding</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=31" title="Edit section: Padding">edit</a>]</div> Padding is the addition of (typically) 0-valued pixels on the borders of an image. This is done so that the border pixels are not undervalued (lost) from the output because they would ordinarily participate in only a single receptive field instance. The padding applied is typically one less than the corresponding kernel dimension. For example, a convolutional layer using 3x3 kernels would receive a 2-pixel pad, that is 1 pixel on each side of the image.[<a href="/wiki/Wikipedia:Citation_needed" title="Wikipedia:Citation needed">citation needed</a>] <div class="mw-heading mw-heading3"><h3 id="Stride">Stride</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=32" title="Edit section: Stride">edit</a>]</div> The stride is the number of pixels that the analysis window moves on each iteration. A stride of 2 means that each kernel is offset by 2 pixels from its predecessor. <div class="mw-heading mw-heading3"><h3 id="Number_of_filters">Number of filters</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=33" title="Edit section: Number of filters">edit</a>]</div> Since feature map size decreases with depth, layers near the input layer tend to have fewer filters while higher layers can have more. To equalize computation at each layer, the product of feature values va with pixel position is kept roughly constant across layers. Preserving more information about the input would require keeping the total number of activations (number of feature maps times number of pixel positions) non-decreasing from one layer to the next. The number of feature maps directly controls the capacity and depends on the number of available examples and task complexity. <div class="mw-heading mw-heading3"><h3 id="Filter_size">Filter size</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=34" title="Edit section: Filter size">edit</a>]</div> Common filter sizes found in the literature vary greatly, and are usually chosen based on the data set. Typical filter sizes range from 1x1 to 7x7. As two famous examples, <a href="/wiki/AlexNet" title="AlexNet">AlexNet</a> used 3x3, 5x5, and 11x11. <a href="/wiki/Inceptionv3" class="mw-redirect" title="Inceptionv3">Inceptionv3</a> used 1x1, 3x3, and 5x5. The challenge is to find the right level of granularity so as to create abstractions at the proper scale, given a particular data set, and without <a href="/wiki/Overfitting" title="Overfitting">overfitting</a>. <div class="mw-heading mw-heading3"><h3 id="Pooling_type_and_size">Pooling type and size</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=35" title="Edit section: Pooling type and size">edit</a>]</div> <a href="/wiki/Max_pooling" class="mw-redirect" title="Max pooling">Max pooling</a> is typically used, often with a 2x2 dimension. This implies that the input is drastically <a href="/wiki/Downsampling_(signal_processing)" title="Downsampling (signal processing)">downsampled</a>, reducing processing cost. Greater pooling <a href="/wiki/Dimensionality_reduction" title="Dimensionality reduction">reduces the dimension</a> of the signal, and may result in unacceptable <a href="/wiki/Data_loss" title="Data loss">information loss</a>. Often, non-overlapping pooling windows perform best.<a href="#cite_note-Scherer-ICANN-2010-78">[76]</a> <div class="mw-heading mw-heading3"><h3 id="Dilation">Dilation</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=36" title="Edit section: Dilation">edit</a>]</div> Dilation involves ignoring pixels within a kernel. This reduces processing/memory potentially without significant signal loss. A dilation of 2 on a 3x3 kernel expands the kernel to 5x5, while still processing 9 (evenly spaced) pixels. Accordingly, dilation of 4 expands the kernel to 7x7.[<a href="/wiki/Wikipedia:Citation_needed" title="Wikipedia:Citation needed">citation needed</a>] <div class="mw-heading mw-heading2"><h2 id="Translation_equivariance_and_aliasing">Translation equivariance and aliasing</h2>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=37" title="Edit section: Translation equivariance and aliasing">edit</a>]</div> It is commonly assumed that CNNs are invariant to shifts of the input. Convolution or pooling layers within a CNN that do not have a stride greater than one are indeed <a href="/wiki/Equivariant_map" title="Equivariant map">equivariant</a> to translations of the input.<a href="#cite_note-:5-72">[72]</a> However, layers with a stride greater than one ignore the <a href="/wiki/Nyquist%E2%80%93Shannon_sampling_theorem" title="Nyquist–Shannon sampling theorem">Nyquist-Shannon sampling theorem</a> and might lead to <a href="/wiki/Aliasing" title="Aliasing">aliasing</a> of the input signal<a href="#cite_note-:5-72">[72]</a> While, in principle, CNNs are capable of implementing anti-aliasing filters, it has been observed that this does not happen in practice <a href="#cite_note-89">[86]</a> and yield models that are not equivariant to translations. Furthermore, if a CNN makes use of fully connected layers, translation equivariance does not imply translation invariance, as the fully connected layers are not invariant to shifts of the input.<a href="#cite_note-90">[87]</a><a href="#cite_note-:6-15">[15]</a> One solution for complete translation invariance is avoiding any down-sampling throughout the network and applying global average pooling at the last layer.<a href="#cite_note-:5-72">[72]</a> Additionally, several other partial solutions have been proposed, such as <a href="/wiki/Anti-aliasing_filter" title="Anti-aliasing filter">anti-aliasing</a> before downsampling operations,<a href="#cite_note-91">[88]</a> spatial transformer networks,<a href="#cite_note-92">[89]</a> <a href="/wiki/Data_augmentation" title="Data augmentation">data augmentation</a>, subsampling combined with pooling,<a href="#cite_note-:6-15">[15]</a> and <a href="/wiki/Capsule_neural_network" title="Capsule neural network">capsule neural networks</a>.<a href="#cite_note-93">[90]</a> <div class="mw-heading mw-heading2"><h2 id="Evaluation">Evaluation</h2>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=38" title="Edit section: Evaluation">edit</a>]</div> The accuracy of the final model is based on a sub-part of the dataset set apart at the start, often called a test-set. Other times methods such as <a href="/wiki/Cross-validation_(statistics)" title="Cross-validation (statistics)">k-fold cross-validation</a> are applied. Other strategies include using <a href="/wiki/Conformal_prediction" title="Conformal prediction">conformal prediction</a>.<a href="#cite_note-94">[91]</a><a href="#cite_note-95">[92]</a> <div class="mw-heading mw-heading2"><h2 id="Regularization_methods">Regularization methods</h2>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=39" title="Edit section: Regularization methods">edit</a>]</div> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1236090951"><div role="note" class="hatnote navigation-not-searchable">Main article: <a href="/wiki/Regularization_(mathematics)" title="Regularization (mathematics)">Regularization (mathematics)</a></div> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1251242444"><table class="box-More_citations_needed_section plainlinks metadata ambox ambox-content ambox-Refimprove" role="presentation"><tbody><tr><td class="mbox-image"><div class="mbox-image-div"><a href="/wiki/File:Question_book-new.svg" class="mw-file-description"><img alt="" src="//upload.wikimedia.org/wikipedia/en/thumb/9/99/Question_book-new.svg/50px-Question_book-new.svg.png" decoding="async" width="50" height="39" class="mw-file-element" srcset="//upload.wikimedia.org/wikipedia/en/thumb/9/99/Question_book-new.svg/75px-Question_book-new.svg.png 1.5x, //upload.wikimedia.org/wikipedia/en/thumb/9/99/Question_book-new.svg/100px-Question_book-new.svg.png 2x" data-file-width="512" data-file-height="399" /></a></div></td><td class="mbox-text"><div class="mbox-text-span">This section needs additional citations for <a href="/wiki/Wikipedia:Verifiability" title="Wikipedia:Verifiability">verification</a>. Please help <a href="/wiki/Special:EditPage/Convolutional_neural_network" title="Special:EditPage/Convolutional neural network">improve this article</a> by <a href="/wiki/Help:Referencing_for_beginners" title="Help:Referencing for beginners">adding citations to reliable sources</a> in this section. Unsourced material may be challenged and removed. (June 2017) (<a href="/wiki/Help:Maintenance_template_removal" title="Help:Maintenance template removal">Learn how and when to remove this message</a>)</div></td></tr></tbody></table> <a href="/wiki/Regularization_(mathematics)" title="Regularization (mathematics)">Regularization</a> is a process of introducing additional information to solve an <a href="/wiki/Ill-posed_problem" class="mw-redirect" title="Ill-posed problem">ill-posed problem</a> or to prevent <a href="/wiki/Overfitting" title="Overfitting">overfitting</a>. CNNs use various types of regularization. <div class="mw-heading mw-heading3"><h3 id="Empirical">Empirical</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=40" title="Edit section: Empirical">edit</a>]</div> <div class="mw-heading mw-heading4"><h4 id="Dropout">Dropout</h4>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=41" title="Edit section: Dropout">edit</a>]</div> Because a fully connected layer occupies most of the parameters, it is prone to overfitting. One method to reduce overfitting is <a href="/wiki/Dropout_(neural_networks)" class="mw-redirect" title="Dropout (neural networks)">dropout</a>, introduced in 2014.<a href="#cite_note-96">[93]</a> At each training stage, individual nodes are either "dropped out" of the net (ignored) with probability <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle 1-p}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mn>1</mn> <mo>−</mo> <mi>p</mi> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle 1-p}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/9633a8692121eedfa99cace406205e5d1511ef8d" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.671ex; width:5.172ex; height:2.509ex;" alt="{\displaystyle 1-p}"> or kept with probability <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle p}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mi>p</mi> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle p}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/81eac1e205430d1f40810df36a0edffdc367af36" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.671ex; margin-left: -0.089ex; width:1.259ex; height:2.009ex;" alt="{\displaystyle p}">, so that a reduced network is left; incoming and outgoing edges to a dropped-out node are also removed. Only the reduced network is trained on the data in that stage. The removed nodes are then reinserted into the network with their original weights. In the training stages, <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle p}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mi>p</mi> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle p}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/81eac1e205430d1f40810df36a0edffdc367af36" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.671ex; margin-left: -0.089ex; width:1.259ex; height:2.009ex;" alt="{\displaystyle p}"> is usually 0.5; for input nodes, it is typically much higher because information is directly lost when input nodes are ignored. At testing time after training has finished, we would ideally like to find a sample average of all possible <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle 2^{n}}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <msup> <mn>2</mn> <mrow class="MJX-TeXAtom-ORD"> <mi>n</mi> </mrow> </msup> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle 2^{n}}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/8226f30650ee4fe4e640c6d2798127e80e9c160d" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.338ex; width:2.381ex; height:2.343ex;" alt="{\displaystyle 2^{n}}"> dropped-out networks; unfortunately this is unfeasible for large values of <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle n}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mi>n</mi> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle n}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/a601995d55609f2d9f5e233e36fbe9ea26011b3b" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.338ex; width:1.395ex; height:1.676ex;" alt="{\displaystyle n}">. However, we can find an approximation by using the full network with each node's output weighted by a factor of <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle p}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mi>p</mi> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle p}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/81eac1e205430d1f40810df36a0edffdc367af36" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.671ex; margin-left: -0.089ex; width:1.259ex; height:2.009ex;" alt="{\displaystyle p}">, so the <a href="/wiki/Expected_value" title="Expected value">expected value</a> of the output of any node is the same as in the training stages. This is the biggest contribution of the dropout method: although it effectively generates <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle 2^{n}}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <msup> <mn>2</mn> <mrow class="MJX-TeXAtom-ORD"> <mi>n</mi> </mrow> </msup> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle 2^{n}}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/8226f30650ee4fe4e640c6d2798127e80e9c160d" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.338ex; width:2.381ex; height:2.343ex;" alt="{\displaystyle 2^{n}}"> neural nets, and as such allows for model combination, at test time only a single network needs to be tested. By avoiding training all nodes on all training data, dropout decreases overfitting. The method also significantly improves training speed. This makes the model combination practical, even for <a href="/wiki/Deep_neural_network" class="mw-redirect" title="Deep neural network">deep neural networks</a>. The technique seems to reduce node interactions, leading them to learn more robust features[<a href="/wiki/Wikipedia:Please_clarify" title="Wikipedia:Please clarify">clarification needed</a>] that better generalize to new data. <div class="mw-heading mw-heading4"><h4 id="DropConnect">DropConnect</h4>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=42" title="Edit section: DropConnect">edit</a>]</div> DropConnect is the generalization of dropout in which each connection, rather than each output unit, can be dropped with probability <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle 1-p}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mn>1</mn> <mo>−</mo> <mi>p</mi> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle 1-p}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/9633a8692121eedfa99cace406205e5d1511ef8d" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.671ex; width:5.172ex; height:2.509ex;" alt="{\displaystyle 1-p}">. Each unit thus receives input from a random subset of units in the previous layer.<a href="#cite_note-97">[94]</a> DropConnect is similar to dropout as it introduces dynamic sparsity within the model, but differs in that the sparsity is on the weights, rather than the output vectors of a layer. In other words, the fully connected layer with DropConnect becomes a sparsely connected layer in which the connections are chosen at random during the training stage. <div class="mw-heading mw-heading4"><h4 id="Stochastic_pooling">Stochastic pooling</h4>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=43" title="Edit section: Stochastic pooling">edit</a>]</div> A major drawback to Dropout is that it does not have the same benefits for convolutional layers, where the neurons are not fully connected. Even before Dropout, in 2013 a technique called stochastic pooling,<a href="#cite_note-98">[95]</a> the conventional <a href="/wiki/Deterministic_algorithm" title="Deterministic algorithm">deterministic</a> pooling operations were replaced with a stochastic procedure, where the activation within each pooling region is picked randomly according to a <a href="/wiki/Multinomial_distribution" title="Multinomial distribution">multinomial distribution</a>, given by the activities within the pooling region. This approach is free of hyperparameters and can be combined with other regularization approaches, such as dropout and <a href="/wiki/Data_augmentation" title="Data augmentation">data augmentation</a>. An alternate view of stochastic pooling is that it is equivalent to standard max pooling but with many copies of an input image, each having small local <a href="/wiki/Deformation_theory" class="mw-redirect" title="Deformation theory">deformations</a>. This is similar to explicit <a href="/wiki/Elastic_deformation" class="mw-redirect" title="Elastic deformation">elastic deformations</a> of the input images,<a href="#cite_note-:3-99">[96]</a> which delivers excellent performance on the <a href="/wiki/MNIST_database" title="MNIST database">MNIST data set</a>.<a href="#cite_note-:3-99">[96]</a> Using stochastic pooling in a multilayer model gives an exponential number of deformations since the selections in higher layers are independent of those below. <div class="mw-heading mw-heading4"><h4 id="Artificial_data">Artificial data</h4>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=44" title="Edit section: Artificial data">edit</a>]</div> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1236090951"><div role="note" class="hatnote navigation-not-searchable">Main article: <a href="/wiki/Data_augmentation" title="Data augmentation">Data augmentation</a></div> Because the degree of model overfitting is determined by both its power and the amount of training it receives, providing a convolutional network with more training examples can reduce overfitting. Because there is often not enough available data to train, especially considering that some part should be spared for later testing, two approaches are to either generate new data from scratch (if possible) or perturb existing data to create new ones. The latter one is used since mid-1990s.<a href="#cite_note-lecun95-52">[52]</a> For example, input images can be cropped, rotated, or rescaled to create new examples with the same labels as the original training set.<a href="#cite_note-100">[97]</a> <div class="mw-heading mw-heading3"><h3 id="Explicit">Explicit</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=45" title="Edit section: Explicit">edit</a>]</div> <div class="mw-heading mw-heading4"><h4 id="Early_stopping">Early stopping</h4>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=46" title="Edit section: Early stopping">edit</a>]</div> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1236090951"><div role="note" class="hatnote navigation-not-searchable">Main article: <a href="/wiki/Early_stopping" title="Early stopping">Early stopping</a></div> One of the simplest methods to prevent overfitting of a network is to simply stop the training before overfitting has had a chance to occur. It comes with the disadvantage that the learning process is halted. <div class="mw-heading mw-heading4"><h4 id="Number_of_parameters">Number of parameters</h4>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=47" title="Edit section: Number of parameters">edit</a>]</div> Another simple way to prevent overfitting is to limit the number of parameters, typically by limiting the number of hidden units in each layer or limiting network depth. For convolutional networks, the filter size also affects the number of parameters. Limiting the number of parameters restricts the predictive power of the network directly, reducing the complexity of the function that it can perform on the data, and thus limits the amount of overfitting. This is equivalent to a "<a href="/wiki/Zero_norm" class="mw-redirect" title="Zero norm">zero norm</a>". <div class="mw-heading mw-heading4"><h4 id="Weight_decay">Weight decay</h4>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=48" title="Edit section: Weight decay">edit</a>]</div> A simple form of added regularizer is weight decay, which simply adds an additional error, proportional to the sum of weights (<a href="/wiki/L1-norm" class="mw-redirect" title="L1-norm">L1 norm</a>) or squared magnitude (<a href="/wiki/L2_norm" class="mw-redirect" title="L2 norm">L2 norm</a>) of the weight vector, to the error at each node. The level of acceptable model complexity can be reduced by increasing the proportionality constant('alpha' hyperparameter), thus increasing the penalty for large weight vectors. L2 regularization is the most common form of regularization. It can be implemented by penalizing the squared magnitude of all parameters directly in the objective. The L2 regularization has the intuitive interpretation of heavily penalizing peaky weight vectors and preferring diffuse weight vectors. Due to multiplicative interactions between weights and inputs this has the useful property of encouraging the network to use all of its inputs a little rather than some of its inputs a lot. L1 regularization is also common. It makes the weight vectors sparse during optimization. In other words, neurons with L1 regularization end up using only a sparse subset of their most important inputs and become nearly invariant to the noisy inputs. L1 with L2 regularization can be combined; this is called <a href="/wiki/Elastic_net_regularization" title="Elastic net regularization">elastic net regularization</a>. <div class="mw-heading mw-heading4"><h4 id="Max_norm_constraints">Max norm constraints</h4>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=49" title="Edit section: Max norm constraints">edit</a>]</div> Another form of regularization is to enforce an absolute upper bound on the magnitude of the weight vector for every neuron and use <a href="/wiki/Sparse_approximation#Projected_Gradient_Descent" title="Sparse approximation">projected gradient descent</a> to enforce the constraint. In practice, this corresponds to performing the parameter update as normal, and then enforcing the constraint by clamping the weight vector <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle {\vec {w}}}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mrow class="MJX-TeXAtom-ORD"> <mrow class="MJX-TeXAtom-ORD"> <mover> <mi>w</mi> <mo stretchy="false">→</mo> </mover> </mrow> </mrow> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle {\vec {w}}}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/8b6c48cdaecf8d81481ea21b1d0c046bf34b68ec" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.338ex; width:1.664ex; height:2.343ex;" alt="{\displaystyle {\vec {w}}}"> of every neuron to satisfy <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle \|{\vec {w}}\|_{2}<c}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mo fence="false" stretchy="false">‖</mo> <mrow class="MJX-TeXAtom-ORD"> <mrow class="MJX-TeXAtom-ORD"> <mover> <mi>w</mi> <mo stretchy="false">→</mo> </mover> </mrow> </mrow> <msub> <mo fence="false" stretchy="false">‖</mo> <mrow class="MJX-TeXAtom-ORD"> <mn>2</mn> </mrow> </msub> <mo><</mo> <mi>c</mi> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle \|{\vec {w}}\|_{2}<c}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/a1a47ec91922a4d2bc7122e90cff40c4b8b73d66" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.838ex; width:9.149ex; height:2.843ex;" alt="{\displaystyle \|{\vec {w}}\|_{2}<c}">. Typical values of <math xmlns="http://www.w3.org/1998/Math/MathML" alttext="{\displaystyle c}"> <semantics> <mrow class="MJX-TeXAtom-ORD"> <mstyle displaystyle="true" scriptlevel="0"> <mi>c</mi> </mstyle> </mrow> <annotation encoding="application/x-tex">{\displaystyle c}</annotation> </semantics> </math><img src="https://wikimedia.org/api/rest_v1/media/math/render/svg/86a67b81c2de995bd608d5b2df50cd8cd7d92455" class="mwe-math-fallback-image-inline mw-invert skin-invert" aria-hidden="true" style="vertical-align: -0.338ex; width:1.007ex; height:1.676ex;" alt="{\displaystyle c}"> are order of 3–4. Some papers report improvements<a href="#cite_note-101">[98]</a> when using this form of regularization. <div class="mw-heading mw-heading2"><h2 id="Hierarchical_coordinate_frames">Hierarchical coordinate frames</h2>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=50" title="Edit section: Hierarchical coordinate frames">edit</a>]</div> Pooling loses the precise spatial relationships between high-level parts (such as nose and mouth in a face image). These relationships are needed for identity recognition. Overlapping the pools so that each feature occurs in multiple pools, helps retain the information. Translation alone cannot extrapolate the understanding of geometric relationships to a radically new viewpoint, such as a different orientation or scale. On the other hand, people are very good at extrapolating; after seeing a new shape once they can recognize it from a different viewpoint.<a href="#cite_note-102">[99]</a> An earlier common way to deal with this problem is to train the network on transformed data in different orientations, scales, lighting, etc. so that the network can cope with these variations. This is computationally intensive for large data-sets. The alternative is to use a hierarchy of coordinate frames and use a group of neurons to represent a conjunction of the shape of the feature and its pose relative to the <a href="/wiki/Retina" title="Retina">retina</a>. The pose relative to the retina is the relationship between the coordinate frame of the retina and the intrinsic features' coordinate frame.<a href="#cite_note-103">[100]</a> Thus, one way to represent something is to embed the coordinate frame within it. This allows large features to be recognized by using the consistency of the poses of their parts (e.g. nose and mouth poses make a consistent prediction of the pose of the whole face). This approach ensures that the higher-level entity (e.g. face) is present when the lower-level (e.g. nose and mouth) agree on its prediction of the pose. The vectors of neuronal activity that represent pose ("pose vectors") allow spatial transformations modeled as linear operations that make it easier for the network to learn the hierarchy of visual entities and generalize across viewpoints. This is similar to the way the human <a href="/wiki/Visual_system" title="Visual system">visual system</a> imposes coordinate frames in order to represent shapes.<a href="#cite_note-104">[101]</a> <div class="mw-heading mw-heading2"><h2 id="Applications">Applications</h2>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=51" title="Edit section: Applications">edit</a>]</div> <div class="mw-heading mw-heading3"><h3 id="Image_recognition">Image recognition</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=52" title="Edit section: Image recognition">edit</a>]</div> CNNs are often used in <a href="/wiki/Image_recognition" class="mw-redirect" title="Image recognition">image recognition</a> systems. In 2012, an <a href="/wiki/Per-comparison_error_rate" title="Per-comparison error rate">error rate</a> of 0.23% on the <a href="/wiki/MNIST_database" title="MNIST database">MNIST database</a> was reported.<a href="#cite_note-mcdns-27">[27]</a> Another paper on using CNN for image classification reported that the learning process was "surprisingly fast"; in the same paper, the best published results as of 2011 were achieved in the MNIST database and the NORB database.<a href="#cite_note-flexible-24">[24]</a> Subsequently, a similar CNN called <a href="/wiki/AlexNet" title="AlexNet">AlexNet</a><a href="#cite_note-quartz-105">[102]</a> won the <a href="/wiki/ImageNet_Large_Scale_Visual_Recognition_Challenge" class="mw-redirect" title="ImageNet Large Scale Visual Recognition Challenge">ImageNet Large Scale Visual Recognition Challenge</a> 2012. When applied to <a href="/wiki/Facial_recognition_system" title="Facial recognition system">facial recognition</a>, CNNs achieved a large decrease in error rate.<a href="#cite_note-106">[103]</a> Another paper reported a 97.6% recognition rate on "5,600 still images of more than 10 subjects".<a href="#cite_note-robust_face_detection-20">[20]</a> CNNs were used to assess <a href="/wiki/Video_quality" title="Video quality">video quality</a> in an objective way after manual training; the resulting system had a very low <a href="/wiki/Root_mean_square_error" class="mw-redirect" title="Root mean square error">root mean square error</a>.<a href="#cite_note-video_quality-107">[104]</a> The <a href="/wiki/ImageNet_Large_Scale_Visual_Recognition_Challenge" class="mw-redirect" title="ImageNet Large Scale Visual Recognition Challenge">ImageNet Large Scale Visual Recognition Challenge</a> is a benchmark in object classification and detection, with millions of images and hundreds of object classes. In the ILSVRC 2014,<a href="#cite_note-ILSVRC2014-108">[105]</a> a large-scale visual recognition challenge, almost every highly ranked team used CNN as their basic framework. The winner <a href="/wiki/GoogLeNet" class="mw-redirect" title="GoogLeNet">GoogLeNet</a><a href="#cite_note-googlenet-109">[106]</a> (the foundation of <a href="/wiki/DeepDream" title="DeepDream">DeepDream</a>) increased the mean average <a href="/wiki/Precision_and_recall" title="Precision and recall">precision</a> of object detection to 0.439329, and reduced classification error to 0.06656, the best result to date. Its network applied more than 30 layers. That performance of convolutional neural networks on the ImageNet tests was close to that of humans.<a href="#cite_note-110">[107]</a> The best algorithms still struggle with objects that are small or thin, such as a small ant on a stem of a flower or a person holding a quill in their hand. They also have trouble with images that have been distorted with filters, an increasingly common phenomenon with modern digital cameras. By contrast, those kinds of images rarely trouble humans. Humans, however, tend to have trouble with other issues. For example, they are not good at classifying objects into fine-grained categories such as the particular breed of dog or species of bird, whereas convolutional neural networks handle this.[<a href="/wiki/Wikipedia:Citation_needed" title="Wikipedia:Citation needed">citation needed</a>] In 2015, a many-layered CNN demonstrated the ability to spot faces from a wide range of angles, including upside down, even when partially occluded, with competitive performance. The network was trained on a database of 200,000 images that included faces at various angles and orientations and a further 20 million images without faces. They used batches of 128 images over 50,000 iterations.<a href="#cite_note-111">[108]</a> <div class="mw-heading mw-heading3"><h3 id="Video_analysis">Video analysis</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=53" title="Edit section: Video analysis">edit</a>]</div> Compared to image data domains, there is relatively little work on applying CNNs to video classification. Video is more complex than images since it has another (temporal) dimension. However, some extensions of CNNs into the video domain have been explored. One approach is to treat space and time as equivalent dimensions of the input and perform convolutions in both time and space.<a href="#cite_note-112">[109]</a><a href="#cite_note-113">[110]</a> Another way is to fuse the features of two convolutional neural networks, one for the spatial and one for the temporal stream.<a href="#cite_note-114">[111]</a><a href="#cite_note-115">[112]</a><a href="#cite_note-116">[113]</a> <a href="/wiki/Long_short-term_memory" title="Long short-term memory">Long short-term memory</a> (LSTM) <a href="/wiki/Recurrent_neural_network" title="Recurrent neural network">recurrent</a> units are typically incorporated after the CNN to account for inter-frame or inter-clip dependencies.<a href="#cite_note-Wang_Duan_Zhang_Niu_p=1657-117">[114]</a><a href="#cite_note-Duan_Wang_Zhai_Zheng_2018_p.-118">[115]</a> <a href="/wiki/Unsupervised_learning" title="Unsupervised learning">Unsupervised learning</a> schemes for training spatio-temporal features have been introduced, based on Convolutional Gated Restricted <a href="/wiki/Boltzmann_machine" title="Boltzmann machine">Boltzmann Machines</a><a href="#cite_note-119">[116]</a> and Independent Subspace Analysis.<a href="#cite_note-120">[117]</a> Its application can be seen in <a href="/wiki/Text-to-video_model" title="Text-to-video model">text-to-video model</a>.[<a href="/wiki/Wikipedia:Citation_needed" title="Wikipedia:Citation needed">citation needed</a>] <div class="mw-heading mw-heading3"><h3 id="Natural_language_processing">Natural language processing</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=54" title="Edit section: Natural language processing">edit</a>]</div> CNNs have also been explored for <a href="/wiki/Natural_language_processing" title="Natural language processing">natural language processing</a>. CNN models are effective for various NLP problems and achieved excellent results in <a href="/wiki/Semantic_parsing" title="Semantic parsing">semantic parsing</a>,<a href="#cite_note-121">[118]</a> search query retrieval,<a href="#cite_note-122">[119]</a> sentence modeling,<a href="#cite_note-123">[120]</a> classification,<a href="#cite_note-124">[121]</a> prediction<a href="#cite_note-125">[122]</a> and other traditional NLP tasks.<a href="#cite_note-126">[123]</a> Compared to traditional language processing methods such as <a href="/wiki/Recurrent_neural_networks" class="mw-redirect" title="Recurrent neural networks">recurrent neural networks</a>, CNNs can represent different contextual realities of language that do not rely on a series-sequence assumption, while RNNs are better suitable when classical time series modeling is required.<a href="#cite_note-127">[124]</a><a href="#cite_note-128">[125]</a><a href="#cite_note-129">[126]</a><a href="#cite_note-130">[127]</a> <div class="mw-heading mw-heading3"><h3 id="Anomaly_detection">Anomaly detection</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=55" title="Edit section: Anomaly detection">edit</a>]</div> A CNN with 1-D convolutions was used on time series in the frequency domain (spectral residual) by an unsupervised model to detect anomalies in the time domain.<a href="#cite_note-131">[128]</a> <div class="mw-heading mw-heading3"><h3 id="Drug_discovery">Drug discovery</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=56" title="Edit section: Drug discovery">edit</a>]</div> CNNs have been used in <a href="/wiki/Drug_discovery" title="Drug discovery">drug discovery</a>. Predicting the interaction between molecules and biological <a href="/wiki/Protein" title="Protein">proteins</a> can identify potential treatments. In 2015, Atomwise introduced AtomNet, the first deep learning neural network for <a href="/wiki/Structure-based_drug_design" class="mw-redirect" title="Structure-based drug design">structure-based drug design</a>.<a href="#cite_note-132">[129]</a> The system trains directly on 3-dimensional representations of chemical interactions. Similar to how image recognition networks learn to compose smaller, spatially proximate features into larger, complex structures,<a href="#cite_note-133">[130]</a> AtomNet discovers chemical features, such as <a href="/wiki/Aromaticity" title="Aromaticity">aromaticity</a>, <a href="/wiki/Orbital_hybridisation" title="Orbital hybridisation">sp3 carbons</a>, and <a href="/wiki/Hydrogen_bond" title="Hydrogen bond">hydrogen bonding</a>. Subsequently, AtomNet was used to predict novel candidate <a href="/wiki/Biomolecule" title="Biomolecule">biomolecules</a> for multiple disease targets, most notably treatments for the <a href="/wiki/Ebola_virus" class="mw-redirect" title="Ebola virus">Ebola virus</a><a href="#cite_note-134">[131]</a> and <a href="/wiki/Multiple_sclerosis" title="Multiple sclerosis">multiple sclerosis</a>.<a href="#cite_note-135">[132]</a> <div class="mw-heading mw-heading3"><h3 id="Checkers_game">Checkers game</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=57" title="Edit section: Checkers game">edit</a>]</div> CNNs have been used in the game of <a href="/wiki/Draughts" class="mw-redirect" title="Draughts">checkers</a>. From 1999 to 2001, <a href="/wiki/David_B._Fogel" title="David B. Fogel">Fogel</a> and Chellapilla published papers showing how a convolutional neural network could learn to play checker using co-evolution. The learning process did not use prior human professional games, but rather focused on a minimal set of information contained in the checkerboard: the location and type of pieces, and the difference in number of pieces between the two sides. Ultimately, the program (<a href="/wiki/Blondie24" title="Blondie24">Blondie24</a>) was tested on 165 games against players and ranked in the highest 0.4%.<a href="#cite_note-136">[133]</a><a href="#cite_note-137">[134]</a> It also earned a win against the program <a href="/wiki/Chinook_(draughts_player)" class="mw-redirect" title="Chinook (draughts player)">Chinook</a> at its "expert" level of play.<a href="#cite_note-138">[135]</a> <div class="mw-heading mw-heading3"><h3 id="Go">Go</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=58" title="Edit section: Go">edit</a>]</div> CNNs have been used in <a href="/wiki/Computer_Go" title="Computer Go">computer Go</a>. In December 2014, Clark and <a href="/wiki/Amos_Storkey" title="Amos Storkey">Storkey</a> published a paper showing that a CNN trained by supervised learning from a database of human professional games could outperform <a href="/wiki/GNU_Go" title="GNU Go">GNU Go</a> and win some games against <a href="/wiki/Monte_Carlo_tree_search" title="Monte Carlo tree search">Monte Carlo tree search</a> Fuego 1.1 in a fraction of the time it took Fuego to play.<a href="#cite_note-139">[136]</a> Later it was announced that a large 12-layer convolutional neural network had correctly predicted the professional move in 55% of positions, equalling the accuracy of a <a href="/wiki/Go_ranks_and_ratings" title="Go ranks and ratings">6 dan</a> human player. When the trained convolutional network was used directly to play games of Go, without any search, it beat the traditional search program GNU Go in 97% of games, and matched the performance of the <a href="/wiki/Monte_Carlo_tree_search" title="Monte Carlo tree search">Monte Carlo tree search</a> program Fuego simulating ten thousand playouts (about a million positions) per move.<a href="#cite_note-140">[137]</a> A couple of CNNs for choosing moves to try ("policy network") and evaluating positions ("value network") driving MCTS were used by <a href="/wiki/AlphaGo" title="AlphaGo">AlphaGo</a>, the first to beat the best human player at the time.<a href="#cite_note-141">[138]</a> <div class="mw-heading mw-heading3"><h3 id="Time_series_forecasting">Time series forecasting</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=59" title="Edit section: Time series forecasting">edit</a>]</div> Recurrent neural networks are generally considered the best neural network architectures for time series forecasting (and sequence modeling in general), but recent studies show that convolutional networks can perform comparably or even better.<a href="#cite_note-142">[139]</a><a href="#cite_note-Tsantekidis_7–12-12">[12]</a> Dilated convolutions<a href="#cite_note-143">[140]</a> might enable one-dimensional convolutional neural networks to effectively learn time series dependences.<a href="#cite_note-144">[141]</a> Convolutions can be implemented more efficiently than RNN-based solutions, and they do not suffer from vanishing (or exploding) gradients.<a href="#cite_note-145">[142]</a> Convolutional networks can provide an improved forecasting performance when there are multiple similar time series to learn from.<a href="#cite_note-146">[143]</a> CNNs can also be applied to further tasks in time series analysis (e.g., time series classification<a href="#cite_note-147">[144]</a> or quantile forecasting<a href="#cite_note-148">[145]</a>). <div class="mw-heading mw-heading3"><h3 id="Cultural_heritage_and_3D-datasets">Cultural heritage and 3D-datasets</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=60" title="Edit section: Cultural heritage and 3D-datasets">edit</a>]</div> As archaeological findings such as <a href="/wiki/Clay_tablet" title="Clay tablet">clay tablets</a> with <a href="/wiki/Cuneiform" title="Cuneiform">cuneiform writing</a> are increasingly acquired using <a href="/wiki/3D_scanner" class="mw-redirect" title="3D scanner">3D scanners</a>, benchmark datasets are becoming available, including HeiCuBeDa<a href="#cite_note-HeiCuBeDa_Hilprecht-149">[146]</a> providing almost 2000 normalized 2-D and 3-D datasets prepared with the <a href="/wiki/GigaMesh_Software_Framework" title="GigaMesh Software Framework">GigaMesh Software Framework</a>.<a href="#cite_note-ICDAR19-150">[147]</a> So <a href="/wiki/Curvature" title="Curvature">curvature</a>-based measures are used in conjunction with geometric neural networks (GNNs), e.g. for period classification of those clay tablets being among the oldest documents of human history.<a href="#cite_note-ICFHR20-151">[148]</a><a href="#cite_note-ICFHR20_Presentation-152">[149]</a> <div class="mw-heading mw-heading2"><h2 id="Fine-tuning">Fine-tuning</h2>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=61" title="Edit section: Fine-tuning">edit</a>]</div> For many applications, training data is not very available. Convolutional neural networks usually require a large amount of training data in order to avoid <a href="/wiki/Overfitting" title="Overfitting">overfitting</a>. A common technique is to train the network on a larger data set from a related domain. Once the network parameters have converged an additional training step is performed using the in-domain data to fine-tune the network weights, this is known as <a href="/wiki/Transfer_learning" title="Transfer learning">transfer learning</a>. Furthermore, this technique allows convolutional network architectures to successfully be applied to problems with tiny training sets.<a href="#cite_note-153">[150]</a> <div class="mw-heading mw-heading2"><h2 id="Human_interpretable_explanations">Human interpretable explanations</h2>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=62" title="Edit section: Human interpretable explanations">edit</a>]</div> End-to-end training and prediction are common practice in <a href="/wiki/Computer_vision" title="Computer vision">computer vision</a>. However, human interpretable explanations are required for <a href="/wiki/Safety-critical_system" title="Safety-critical system">critical systems</a> such as a <a href="/wiki/Self-driving_car" title="Self-driving car">self-driving cars</a>.<a href="#cite_note-Interpretable_ML_Symposium_2017-154">[151]</a> With recent advances in <a href="/wiki/Salience_(neuroscience)" title="Salience (neuroscience)">visual salience</a>, <a href="/wiki/Visual_spatial_attention" title="Visual spatial attention">spatial attention</a>, and <a href="/wiki/Visual_temporal_attention" title="Visual temporal attention">temporal attention</a>, the most critical spatial regions/temporal instants could be visualized to justify the CNN predictions.<a href="#cite_note-Zang_Wang_Liu_Zhang_2018_pp._97–108-155">[152]</a><a href="#cite_note-Wang_Zang_Zhang_Niu_p=1979-156">[153]</a> <div class="mw-heading mw-heading2"><h2 id="Related_architectures">Related architectures</h2>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=63" title="Edit section: Related architectures">edit</a>]</div> <div class="mw-heading mw-heading3"><h3 id="Deep_Q-networks">Deep Q-networks</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=64" title="Edit section: Deep Q-networks">edit</a>]</div> A deep Q-network (DQN) is a type of deep learning model that combines a deep neural network with <a href="/wiki/Q-learning" title="Q-learning">Q-learning</a>, a form of <a href="/wiki/Reinforcement_learning" title="Reinforcement learning">reinforcement learning</a>. Unlike earlier reinforcement learning agents, DQNs that utilize CNNs can learn directly from high-dimensional sensory inputs via reinforcement learning.<a href="#cite_note-Ong_Chavez_Hong_2015-157">[154]</a> Preliminary results were presented in 2014, with an accompanying paper in February 2015.<a href="#cite_note-DQN-158">[155]</a> The research described an application to <a href="/wiki/Atari_2600" title="Atari 2600">Atari 2600</a> gaming. Other deep reinforcement learning models preceded it.<a href="#cite_note-159">[156]</a> <div class="mw-heading mw-heading3"><h3 id="Deep_belief_networks">Deep belief networks</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=65" title="Edit section: Deep belief networks">edit</a>]</div> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1236090951"><div role="note" class="hatnote navigation-not-searchable">Main article: <a href="/wiki/Deep_belief_network" title="Deep belief network">Deep belief network</a></div> <a href="/wiki/Convolutional_deep_belief_network" title="Convolutional deep belief network">Convolutional deep belief networks</a> (CDBN) have structure very similar to convolutional neural networks and are trained similarly to deep belief networks. Therefore, they exploit the 2D structure of images, like CNNs do, and make use of pre-training like <a href="/wiki/Deep_belief_network" title="Deep belief network">deep belief networks</a>. They provide a generic structure that can be used in many image and signal processing tasks. Benchmark results on standard image datasets like CIFAR<a href="#cite_note-CDBN-CIFAR-160">[157]</a> have been obtained using CDBNs.<a href="#cite_note-CDBN-161">[158]</a><figure class="mw-default-size" typeof="mw:File/Thumb"><a href="/wiki/File:Neural_Abstraction_Pyramid.jpg" class="mw-file-description"><img alt="Neural Abstraction Pyramid" src="//upload.wikimedia.org/wikipedia/commons/thumb/d/d9/Neural_Abstraction_Pyramid.jpg/220px-Neural_Abstraction_Pyramid.jpg" decoding="async" width="220" height="128" class="mw-file-element" srcset="//upload.wikimedia.org/wikipedia/commons/thumb/d/d9/Neural_Abstraction_Pyramid.jpg/330px-Neural_Abstraction_Pyramid.jpg 1.5x, //upload.wikimedia.org/wikipedia/commons/thumb/d/d9/Neural_Abstraction_Pyramid.jpg/440px-Neural_Abstraction_Pyramid.jpg 2x" data-file-width="1343" data-file-height="782" /></a><figcaption>Neural abstraction pyramid</figcaption></figure> <div class="mw-heading mw-heading3"><h3 id="Neural_abstraction_pyramid">Neural abstraction pyramid</h3>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=66" title="Edit section: Neural abstraction pyramid">edit</a>]</div> The feed-forward architecture of convolutional neural networks was extended in the neural abstraction pyramid<a href="#cite_note-162">[159]</a> by lateral and feedback connections. The resulting recurrent convolutional network allows for the flexible incorporation of contextual information to iteratively resolve local ambiguities. In contrast to previous models, image-like outputs at the highest resolution were generated, e.g., for semantic segmentation, image reconstruction, and object localization tasks. <div class="mw-heading mw-heading2"><h2 id="Notable_libraries">Notable libraries</h2>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=67" title="Edit section: Notable libraries">edit</a>]</div> <ul><li><a href="/wiki/Caffe_(software)" title="Caffe (software)">Caffe</a>: A library for convolutional neural networks. Created by the Berkeley Vision and Learning Center (BVLC). It supports both CPU and GPU. Developed in <a href="/wiki/C%2B%2B" title="C++">C++</a>, and has <a href="/wiki/Python_(programming_language)" title="Python (programming language)">Python</a> and <a href="/wiki/MATLAB" title="MATLAB">MATLAB</a> wrappers.</li> <li><a href="/wiki/Deeplearning4j" title="Deeplearning4j">Deeplearning4j</a>: Deep learning in <a href="/wiki/Java_(programming_language)" title="Java (programming language)">Java</a> and <a href="/wiki/Scala_(programming_language)" title="Scala (programming language)">Scala</a> on multi-GPU-enabled <a href="/wiki/Apache_Spark" title="Apache Spark">Spark</a>. A general-purpose deep learning library for the JVM production stack running on a C++ scientific computing engine. Allows the creation of custom layers. Integrates with Hadoop and Kafka.</li> <li><a href="/wiki/Dlib" title="Dlib">Dlib</a>: A toolkit for making real world machine learning and data analysis applications in C++.</li> <li><a href="/wiki/Microsoft_Cognitive_Toolkit" title="Microsoft Cognitive Toolkit">Microsoft Cognitive Toolkit</a>: A deep learning toolkit written by Microsoft with several unique features enhancing scalability over multiple nodes. It supports full-fledged interfaces for training in C++ and Python and with additional support for model inference in <a href="/wiki/C_Sharp_(programming_language)" title="C Sharp (programming language)">C#</a> and Java.</li> <li><a href="/wiki/TensorFlow" title="TensorFlow">TensorFlow</a>: <a href="/wiki/Apache_License#Version_2.0" title="Apache License">Apache 2.0</a>-licensed Theano-like library with support for CPU, GPU, Google's proprietary <a href="/wiki/Tensor_processing_unit" class="mw-redirect" title="Tensor processing unit">tensor processing unit</a> (TPU),<a href="#cite_note-163">[160]</a> and mobile devices.</li> <li><a href="/wiki/Theano_(software)" title="Theano (software)">Theano</a>: The reference deep-learning library for Python with an API largely compatible with the popular <a href="/wiki/NumPy" title="NumPy">NumPy</a> library. Allows user to write symbolic mathematical expressions, then automatically generates their derivatives, saving the user from having to code gradients or backpropagation. These symbolic expressions are automatically compiled to <a href="/wiki/CUDA" title="CUDA">CUDA</a> code for a fast, <a href="/wiki/Compute_kernel" title="Compute kernel">on-the-GPU</a> implementation.</li> <li><a href="/wiki/Torch_(machine_learning)" title="Torch (machine learning)">Torch</a>: A <a href="/wiki/Scientific_computing" class="mw-redirect" title="Scientific computing">scientific computing</a> framework with wide support for machine learning algorithms, written in <a href="/wiki/C_(programming_language)" title="C (programming language)">C</a> and <a href="/wiki/Lua_(programming_language)" title="Lua (programming language)">Lua</a>.</li></ul> <div class="mw-heading mw-heading2"><h2 id="See_also">See also</h2>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=68" title="Edit section: See also">edit</a>]</div> <ul><li><a href="/wiki/Attention_(machine_learning)" title="Attention (machine learning)">Attention (machine learning)</a></li> <li><a href="/wiki/Convolution" title="Convolution">Convolution</a></li> <li><a href="/wiki/Deep_learning" title="Deep learning">Deep learning</a></li> <li><a href="/wiki/Natural-language_processing" class="mw-redirect" title="Natural-language processing">Natural-language processing</a></li> <li><a href="/wiki/Neocognitron" title="Neocognitron">Neocognitron</a></li> <li><a href="/wiki/Scale-invariant_feature_transform" title="Scale-invariant feature transform">Scale-invariant feature transform</a></li> <li><a href="/wiki/Time_delay_neural_network" title="Time delay neural network">Time delay neural network</a></li> <li><a href="/wiki/Vision_processing_unit" title="Vision processing unit">Vision processing unit</a></li></ul> <div class="mw-heading mw-heading2"><h2 id="Notes">Notes</h2>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=69" title="Edit section: Notes">edit</a>]</div> <style data-mw-deduplicate="TemplateStyles:r1239543626">.mw-parser-output .reflist{margin-bottom:0.5em;list-style-type:decimal}@media screen{.mw-parser-output .reflist{font-size:90%}}.mw-parser-output .reflist .references{font-size:100%;margin-bottom:0;list-style-type:inherit}.mw-parser-output .reflist-columns-2{column-width:30em}.mw-parser-output .reflist-columns-3{column-width:25em}.mw-parser-output .reflist-columns{margin-top:0.3em}.mw-parser-output .reflist-columns ol{margin-top:0}.mw-parser-output .reflist-columns li{page-break-inside:avoid;break-inside:avoid-column}.mw-parser-output .reflist-upper-alpha{list-style-type:upper-alpha}.mw-parser-output .reflist-upper-roman{list-style-type:upper-roman}.mw-parser-output .reflist-lower-alpha{list-style-type:lower-alpha}.mw-parser-output .reflist-lower-greek{list-style-type:lower-greek}.mw-parser-output .reflist-lower-roman{list-style-type:lower-roman}</style><div class="reflist"> <div class="mw-references-wrap"><ol class="references"> <li id="cite_note-74"><a href="#cite_ref-74">^</a> When applied to other types of data than image data, such as sound data, "spatial position" may variously correspond to different points in the <a href="/wiki/Time_domain" title="Time domain">time domain</a>, <a href="/wiki/Frequency_domain" title="Frequency domain">frequency domain</a>, or other <a href="/wiki/Space_(mathematics)" title="Space (mathematics)">mathematical spaces</a>. </li> <li id="cite_note-76"><a href="#cite_ref-76">^</a> hence the name "convolutional layer" </li> <li id="cite_note-88"><a href="#cite_ref-88">^</a> So-called <a href="/wiki/Categorical_data" class="mw-redirect" title="Categorical data">categorical data</a>. </li> </ol></div></div> <div class="mw-heading mw-heading2"><h2 id="References">References</h2>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=70" title="Edit section: References">edit</a>]</div> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1239543626"><div class="reflist reflist-columns references-column-width" style="column-width: 30em;"> <ol class="references"> <li id="cite_note-1"><a href="#cite_ref-1">^</a> <style data-mw-deduplicate="TemplateStyles:r1238218222">.mw-parser-output cite.citation{font-style:inherit;word-wrap:break-word}.mw-parser-output .citation q{quotes:"\"""\"""'""'"}.mw-parser-output .citation:target{background-color:rgba(0,127,255,0.133)}.mw-parser-output .id-lock-free.id-lock-free a{background:url("//upload.wikimedia.org/wikipedia/commons/6/65/Lock-green.svg")right 0.1em center/9px no-repeat}.mw-parser-output .id-lock-limited.id-lock-limited a,.mw-parser-output .id-lock-registration.id-lock-registration a{background:url("//upload.wikimedia.org/wikipedia/commons/d/d6/Lock-gray-alt-2.svg")right 0.1em center/9px no-repeat}.mw-parser-output .id-lock-subscription.id-lock-subscription a{background:url("//upload.wikimedia.org/wikipedia/commons/a/aa/Lock-red-alt-2.svg")right 0.1em center/9px no-repeat}.mw-parser-output .cs1-ws-icon a{background:url("//upload.wikimedia.org/wikipedia/commons/4/4c/Wikisource-logo.svg")right 0.1em center/12px no-repeat}body:not(.skin-timeless):not(.skin-minerva) .mw-parser-output .id-lock-free a,body:not(.skin-timeless):not(.skin-minerva) .mw-parser-output .id-lock-limited a,body:not(.skin-timeless):not(.skin-minerva) .mw-parser-output .id-lock-registration a,body:not(.skin-timeless):not(.skin-minerva) .mw-parser-output .id-lock-subscription a,body:not(.skin-timeless):not(.skin-minerva) .mw-parser-output .cs1-ws-icon a{background-size:contain;padding:0 1em 0 0}.mw-parser-output .cs1-code{color:inherit;background:inherit;border:none;padding:inherit}.mw-parser-output .cs1-hidden-error{display:none;color:var(--color-error,#d33)}.mw-parser-output .cs1-visible-error{color:var(--color-error,#d33)}.mw-parser-output .cs1-maint{display:none;color:#085;margin-left:0.3em}.mw-parser-output .cs1-kern-left{padding-left:0.2em}.mw-parser-output .cs1-kern-right{padding-right:0.2em}.mw-parser-output .citation .mw-selflink{font-weight:inherit}@media screen{.mw-parser-output .cs1-format{font-size:95%}html.skin-theme-clientpref-night .mw-parser-output .cs1-maint{color:#18911f}}@media screen and (prefers-color-scheme:dark){html.skin-theme-clientpref-os .mw-parser-output .cs1-maint{color:#18911f}}</style><cite id="CITEREFLeCunBengioHinton2015" class="citation journal cs1">LeCun, Yann; Bengio, Yoshua; Hinton, Geoffrey (2015-05-28). <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/26017442/">"Deep learning"</a>. Nature. 521 (7553): 436–444. <a href="/wiki/Bibcode_(identifier)" class="mw-redirect" title="Bibcode (identifier)">Bibcode</a>:<a rel="nofollow" class="external text" href="https://ui.adsabs.harvard.edu/abs/2015Natur.521..436L">2015Natur.521..436L</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1038%2Fnature14539">10.1038/nature14539</a>. <a href="/wiki/ISSN_(identifier)" class="mw-redirect" title="ISSN (identifier)">ISSN</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/issn/1476-4687">1476-4687</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/26017442">26017442</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Nature&rft.atitle=Deep+learning&rft.volume=521&rft.issue=7553&rft.pages=436-444&rft.date=2015-05-28&rft_id=info%3Adoi%2F10.1038%2Fnature14539&rft.issn=1476-4687&rft_id=info%3Apmid%2F26017442&rft_id=info%3Abibcode%2F2015Natur.521..436L&rft.aulast=LeCun&rft.aufirst=Yann&rft.au=Bengio%2C+Yoshua&rft.au=Hinton%2C+Geoffrey&rft_id=https%3A%2F%2Fpubmed.ncbi.nlm.nih.gov%2F26017442%2F&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-auto3-2">^ <a href="#cite_ref-auto3_2-0">a</a> <a href="#cite_ref-auto3_2-1">b</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFVenkatesanLi2017" class="citation book cs1">Venkatesan, Ragav; Li, Baoxin (2017-10-23). <a rel="nofollow" class="external text" href="https://books.google.com/books?id=bAM7DwAAQBAJ&q=vanishing+gradient">Convolutional Neural Networks in Visual Computing: A Concise Guide</a>. CRC Press. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-1-351-65032-8" title="Special:BookSources/978-1-351-65032-8"><bdi>978-1-351-65032-8</bdi></a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20231016190415/https://books.google.com/books?id=bAM7DwAAQBAJ&q=vanishing+gradient#v=snippet&q=vanishing%20gradient&f=false">Archived</a> from the original on 2023-10-16. Retrieved 2020-12-13.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=book&rft.btitle=Convolutional+Neural+Networks+in+Visual+Computing%3A+A+Concise+Guide&rft.pub=CRC+Press&rft.date=2017-10-23&rft.isbn=978-1-351-65032-8&rft.aulast=Venkatesan&rft.aufirst=Ragav&rft.au=Li%2C+Baoxin&rft_id=https%3A%2F%2Fbooks.google.com%2Fbooks%3Fid%3DbAM7DwAAQBAJ%26q%3Dvanishing%2Bgradient&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-auto2-3">^ <a href="#cite_ref-auto2_3-0">a</a> <a href="#cite_ref-auto2_3-1">b</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFBalasKumarSrivastava2019" class="citation book cs1">Balas, Valentina E.; Kumar, Raghvendra; Srivastava, Rajshree (2019-11-19). <a rel="nofollow" class="external text" href="https://books.google.com/books?id=XRS_DwAAQBAJ&q=exploding+gradient">Recent Trends and Advances in Artificial Intelligence and Internet of Things</a>. Springer Nature. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-3-030-32644-9" title="Special:BookSources/978-3-030-32644-9"><bdi>978-3-030-32644-9</bdi></a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20231016190414/https://books.google.com/books?id=XRS_DwAAQBAJ&q=exploding+gradient#v=snippet&q=exploding%20gradient&f=false">Archived</a> from the original on 2023-10-16. Retrieved 2020-12-13.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=book&rft.btitle=Recent+Trends+and+Advances+in+Artificial+Intelligence+and+Internet+of+Things&rft.pub=Springer+Nature&rft.date=2019-11-19&rft.isbn=978-3-030-32644-9&rft.aulast=Balas&rft.aufirst=Valentina+E.&rft.au=Kumar%2C+Raghvendra&rft.au=Srivastava%2C+Rajshree&rft_id=https%3A%2F%2Fbooks.google.com%2Fbooks%3Fid%3DXRS_DwAAQBAJ%26q%3Dexploding%2Bgradient&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-4"><a href="#cite_ref-4">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFZhangSoonYeFuh2020" class="citation journal cs1">Zhang, Yingjie; Soon, Hong Geok; Ye, Dongsen; Fuh, Jerry Ying Hsi; Zhu, Kunpeng (September 2020). <a rel="nofollow" class="external text" href="https://ieeexplore.ieee.org/document/8913613">"Powder-Bed Fusion Process Monitoring by Machine Vision With Hybrid Convolutional Neural Networks"</a>. IEEE Transactions on Industrial Informatics. 16 (9): 5769–5779. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2FTII.2019.2956078">10.1109/TII.2019.2956078</a>. <a href="/wiki/ISSN_(identifier)" class="mw-redirect" title="ISSN (identifier)">ISSN</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/issn/1941-0050">1941-0050</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:213010088">213010088</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20230731120013/https://ieeexplore.ieee.org/document/8913613/">Archived</a> from the original on 2023-07-31. Retrieved 2023-08-12.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=IEEE+Transactions+on+Industrial+Informatics&rft.atitle=Powder-Bed+Fusion+Process+Monitoring+by+Machine+Vision+With+Hybrid+Convolutional+Neural+Networks&rft.volume=16&rft.issue=9&rft.pages=5769-5779&rft.date=2020-09&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A213010088%23id-name%3DS2CID&rft.issn=1941-0050&rft_id=info%3Adoi%2F10.1109%2FTII.2019.2956078&rft.aulast=Zhang&rft.aufirst=Yingjie&rft.au=Soon%2C+Hong+Geok&rft.au=Ye%2C+Dongsen&rft.au=Fuh%2C+Jerry+Ying+Hsi&rft.au=Zhu%2C+Kunpeng&rft_id=https%3A%2F%2Fieeexplore.ieee.org%2Fdocument%2F8913613&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-5"><a href="#cite_ref-5">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFChervyakovLyakhovDeryabinNagornov2020" class="citation journal cs1">Chervyakov, N.I.; Lyakhov, P.A.; Deryabin, M.A.; Nagornov, N.N.; Valueva, M.V.; Valuev, G.V. (September 2020). <a rel="nofollow" class="external text" href="https://linkinghub.elsevier.com/retrieve/pii/S092523122030583X">"Residue Number System-Based Solution for Reducing the Hardware Cost of a Convolutional Neural Network"</a>. Neurocomputing. 407: 439–453. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1016%2Fj.neucom.2020.04.018">10.1016/j.neucom.2020.04.018</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:219470398">219470398</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20230629155646/https://linkinghub.elsevier.com/retrieve/pii/S092523122030583X">Archived</a> from the original on 2023-06-29. Retrieved 2023-08-12. <q>Convolutional neural networks represent deep learning architectures that are currently used in a wide range of applications, including computer vision, speech recognition, malware dedection, time series analysis in finance, and many others.</q></cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Neurocomputing&rft.atitle=Residue+Number+System-Based+Solution+for+Reducing+the+Hardware+Cost+of+a+Convolutional+Neural+Network&rft.volume=407&rft.pages=439-453&rft.date=2020-09&rft_id=info%3Adoi%2F10.1016%2Fj.neucom.2020.04.018&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A219470398%23id-name%3DS2CID&rft.aulast=Chervyakov&rft.aufirst=N.I.&rft.au=Lyakhov%2C+P.A.&rft.au=Deryabin%2C+M.A.&rft.au=Nagornov%2C+N.N.&rft.au=Valueva%2C+M.V.&rft.au=Valuev%2C+G.V.&rft_id=https%3A%2F%2Flinkinghub.elsevier.com%2Fretrieve%2Fpii%2FS092523122030583X&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-auto1-6">^ <a href="#cite_ref-auto1_6-0">a</a> <a href="#cite_ref-auto1_6-1">b</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFHabibi2017" class="citation book cs1">Habibi, Aghdam, Hamed (2017-05-30). Guide to convolutional neural networks : a practical application to traffic-sign detection and classification. Heravi, Elnaz Jahani. Cham, Switzerland. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/9783319575490" title="Special:BookSources/9783319575490"><bdi>9783319575490</bdi></a>. <a href="/wiki/OCLC_(identifier)" class="mw-redirect" title="OCLC (identifier)">OCLC</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/oclc/987790957">987790957</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=book&rft.btitle=Guide+to+convolutional+neural+networks+%3A+a+practical+application+to+traffic-sign+detection+and+classification&rft.place=Cham%2C+Switzerland&rft.date=2017-05-30&rft_id=info%3Aoclcnum%2F987790957&rft.isbn=9783319575490&rft.aulast=Habibi&rft.aufirst=Aghdam%2C+Hamed&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"><code class="cs1-code">{{<a href="/wiki/Template:Cite_book" title="Template:Cite book">cite book</a>}}</code>: CS1 maint: location missing publisher (<a href="/wiki/Category:CS1_maint:_location_missing_publisher" title="Category:CS1 maint: location missing publisher">link</a>) CS1 maint: multiple names: authors list (<a href="/wiki/Category:CS1_maint:_multiple_names:_authors_list" title="Category:CS1 maint: multiple names: authors list">link</a>) </li> <li id="cite_note-homma-7">^ <a href="#cite_ref-homma_7-0">a</a> <a href="#cite_ref-homma_7-1">b</a> <a href="#cite_ref-homma_7-2">c</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFHommaLes_AtlasRobert_Marks_II1987" class="citation journal cs1">Homma, Toshiteru; Les Atlas; Robert Marks II (1987). <a rel="nofollow" class="external text" href="https://proceedings.neurips.cc/paper/1987/file/98f13708210194c475687be6106a3b84-Paper.pdf">"An Artificial Neural Network for Spatio-Temporal Bipolar Patterns: Application to Phoneme Classification"</a> (PDF). Advances in Neural Information Processing Systems. 1: 31–40. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20220331211142/https://proceedings.neurips.cc/paper/1987/file/98f13708210194c475687be6106a3b84-Paper.pdf">Archived</a> (PDF) from the original on 2022-03-31. Retrieved 2022-03-31. <q>The notion of convolution or correlation used in the models presented is popular in engineering disciplines and has been applied extensively to designing filters, control systems, etc.</q></cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Advances+in+Neural+Information+Processing+Systems&rft.atitle=An+Artificial+Neural+Network+for+Spatio-Temporal+Bipolar+Patterns%3A+Application+to+Phoneme+Classification&rft.volume=1&rft.pages=31-40&rft.date=1987&rft.aulast=Homma&rft.aufirst=Toshiteru&rft.au=Les+Atlas&rft.au=Robert+Marks+II&rft_id=https%3A%2F%2Fproceedings.neurips.cc%2Fpaper%2F1987%2Ffile%2F98f13708210194c475687be6106a3b84-Paper.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-Valueva_Nagornov_Lyakhov_Valuev_2020_pp._232–243-8"><a href="#cite_ref-Valueva_Nagornov_Lyakhov_Valuev_2020_pp._232–243_8-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFValuevaNagornovLyakhovValuev2020" class="citation journal cs1">Valueva, M.V.; Nagornov, N.N.; Lyakhov, P.A.; Valuev, G.V.; Chervyakov, N.I. (2020). "Application of the residue number system to reduce hardware costs of the convolutional neural network implementation". Mathematics and Computers in Simulation. 177. Elsevier BV: 232–243. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1016%2Fj.matcom.2020.04.031">10.1016/j.matcom.2020.04.031</a>. <a href="/wiki/ISSN_(identifier)" class="mw-redirect" title="ISSN (identifier)">ISSN</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/issn/0378-4754">0378-4754</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:218955622">218955622</a>. <q>Convolutional neural networks are a promising tool for solving the problem of pattern recognition.</q></cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Mathematics+and+Computers+in+Simulation&rft.atitle=Application+of+the+residue+number+system+to+reduce+hardware+costs+of+the+convolutional+neural+network+implementation&rft.volume=177&rft.pages=232-243&rft.date=2020&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A218955622%23id-name%3DS2CID&rft.issn=0378-4754&rft_id=info%3Adoi%2F10.1016%2Fj.matcom.2020.04.031&rft.aulast=Valueva&rft.aufirst=M.V.&rft.au=Nagornov%2C+N.N.&rft.au=Lyakhov%2C+P.A.&rft.au=Valuev%2C+G.V.&rft.au=Chervyakov%2C+N.I.&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-9"><a href="#cite_ref-9">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFvan_den_OordDielemanSchrauwen2013" class="citation book cs1">van den Oord, Aaron; Dieleman, Sander; Schrauwen, Benjamin (2013-01-01). Burges, C. J. C.; Bottou, L.; Welling, M.; Ghahramani, Z.; Weinberger, K. Q. (eds.). <a rel="nofollow" class="external text" href="https://proceedings.neurips.cc/paper/2013/file/b3ba8f1bee1238a2f37603d90b58898d-Paper.pdf">Deep content-based music recommendation</a> (PDF). Curran Associates, Inc. pp. 2643–2651. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20220307172303/https://proceedings.neurips.cc/paper/2013/file/b3ba8f1bee1238a2f37603d90b58898d-Paper.pdf">Archived</a> (PDF) from the original on 2022-03-07. Retrieved 2022-03-31.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=book&rft.btitle=Deep+content-based+music+recommendation&rft.pages=2643-2651&rft.pub=Curran+Associates%2C+Inc.&rft.date=2013-01-01&rft.aulast=van+den+Oord&rft.aufirst=Aaron&rft.au=Dieleman%2C+Sander&rft.au=Schrauwen%2C+Benjamin&rft_id=https%3A%2F%2Fproceedings.neurips.cc%2Fpaper%2F2013%2Ffile%2Fb3ba8f1bee1238a2f37603d90b58898d-Paper.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-10"><a href="#cite_ref-10">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFCollobertWeston2008" class="citation book cs1">Collobert, Ronan; Weston, Jason (2008-01-01). "A unified architecture for natural language processing". Proceedings of the 25th international conference on Machine learning - ICML '08. New York, NY, US: ACM. pp. 160–167. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1145%2F1390156.1390177">10.1145/1390156.1390177</a>. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-1-60558-205-4" title="Special:BookSources/978-1-60558-205-4"><bdi>978-1-60558-205-4</bdi></a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:2617020">2617020</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=bookitem&rft.atitle=A+unified+architecture+for+natural+language+processing&rft.btitle=Proceedings+of+the+25th+international+conference+on+Machine+learning+-+ICML+%2708&rft.place=New+York%2C+NY%2C+US&rft.pages=160-167&rft.pub=ACM&rft.date=2008-01-01&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A2617020%23id-name%3DS2CID&rft_id=info%3Adoi%2F10.1145%2F1390156.1390177&rft.isbn=978-1-60558-205-4&rft.aulast=Collobert&rft.aufirst=Ronan&rft.au=Weston%2C+Jason&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-11"><a href="#cite_ref-11">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFAvilovRimbertPopovBougrain2020" class="citation book cs1">Avilov, Oleksii; Rimbert, Sebastien; Popov, Anton; Bougrain, Laurent (July 2020). <a rel="nofollow" class="external text" href="https://ieeexplore.ieee.org/document/9176228">"Deep Learning Techniques to Improve Intraoperative Awareness Detection from Electroencephalographic Signals"</a>. <a rel="nofollow" class="external text" href="https://hal.inria.fr/hal-02920320/file/Avilov_EMBC2020.pdf">2020 42nd Annual International Conference of the IEEE Engineering in Medicine & Biology Society (EMBC)</a> (PDF). Vol. 2020. Montreal, QC, Canada: IEEE. pp. 142–145. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2FEMBC44109.2020.9176228">10.1109/EMBC44109.2020.9176228</a>. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-1-7281-1990-8" title="Special:BookSources/978-1-7281-1990-8"><bdi>978-1-7281-1990-8</bdi></a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/33017950">33017950</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:221386616">221386616</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20220519135428/https://hal.inria.fr/hal-02920320/file/Avilov_EMBC2020.pdf">Archived</a> (PDF) from the original on 2022-05-19. Retrieved 2023-07-21.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=bookitem&rft.atitle=Deep+Learning+Techniques+to+Improve+Intraoperative+Awareness+Detection+from+Electroencephalographic+Signals&rft.btitle=2020+42nd+Annual+International+Conference+of+the+IEEE+Engineering+in+Medicine+%26+Biology+Society+%28EMBC%29&rft.place=Montreal%2C+QC%2C+Canada&rft.pages=142-145&rft.pub=IEEE&rft.date=2020-07&rft_id=info%3Adoi%2F10.1109%2FEMBC44109.2020.9176228&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A221386616%23id-name%3DS2CID&rft_id=info%3Apmid%2F33017950&rft.isbn=978-1-7281-1990-8&rft.aulast=Avilov&rft.aufirst=Oleksii&rft.au=Rimbert%2C+Sebastien&rft.au=Popov%2C+Anton&rft.au=Bougrain%2C+Laurent&rft_id=https%3A%2F%2Fieeexplore.ieee.org%2Fdocument%2F9176228&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-Tsantekidis_7–12-12">^ <a href="#cite_ref-Tsantekidis_7–12_12-0">a</a> <a href="#cite_ref-Tsantekidis_7–12_12-1">b</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFTsantekidisPassalisTefasKanniainen2017" class="citation book cs1">Tsantekidis, Avraam; Passalis, Nikolaos; Tefas, Anastasios; Kanniainen, Juho; Gabbouj, Moncef; Iosifidis, Alexandros (July 2017). "Forecasting Stock Prices from the Limit Order Book Using Convolutional Neural Networks". 2017 IEEE 19th Conference on Business Informatics (CBI). Thessaloniki, Greece: IEEE. pp. 7–12. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2FCBI.2017.23">10.1109/CBI.2017.23</a>. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-1-5386-3035-8" title="Special:BookSources/978-1-5386-3035-8"><bdi>978-1-5386-3035-8</bdi></a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:4950757">4950757</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=bookitem&rft.atitle=Forecasting+Stock+Prices+from+the+Limit+Order+Book+Using+Convolutional+Neural+Networks&rft.btitle=2017+IEEE+19th+Conference+on+Business+Informatics+%28CBI%29&rft.place=Thessaloniki%2C+Greece&rft.pages=7-12&rft.pub=IEEE&rft.date=2017-07&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A4950757%23id-name%3DS2CID&rft_id=info%3Adoi%2F10.1109%2FCBI.2017.23&rft.isbn=978-1-5386-3035-8&rft.aulast=Tsantekidis&rft.aufirst=Avraam&rft.au=Passalis%2C+Nikolaos&rft.au=Tefas%2C+Anastasios&rft.au=Kanniainen%2C+Juho&rft.au=Gabbouj%2C+Moncef&rft.au=Iosifidis%2C+Alexandros&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-:0-13">^ <a href="#cite_ref-:0_13-0">a</a> <a href="#cite_ref-:0_13-1">b</a> <a href="#cite_ref-:0_13-2">c</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFZhang1988" class="citation journal cs1">Zhang, Wei (1988). <a rel="nofollow" class="external text" href="https://drive.google.com/file/d/1nN_5odSG_QVae54EsQN_qSz-0ZsX6wA0/view?usp=sharing">"Shift-invariant pattern recognition neural network and its optical architecture"</a>. Proceedings of Annual Conference of the Japan Society of Applied Physics. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20200623051222/https://drive.google.com/file/d/1nN_5odSG_QVae54EsQN_qSz-0ZsX6wA0/view?usp=sharing">Archived</a> from the original on 2020-06-23. Retrieved 2020-06-22.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Proceedings+of+Annual+Conference+of+the+Japan+Society+of+Applied+Physics&rft.atitle=Shift-invariant+pattern+recognition+neural+network+and+its+optical+architecture&rft.date=1988&rft.aulast=Zhang&rft.aufirst=Wei&rft_id=https%3A%2F%2Fdrive.google.com%2Ffile%2Fd%2F1nN_5odSG_QVae54EsQN_qSz-0ZsX6wA0%2Fview%3Fusp%3Dsharing&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-:1-14">^ <a href="#cite_ref-:1_14-0">a</a> <a href="#cite_ref-:1_14-1">b</a> <a href="#cite_ref-:1_14-2">c</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFZhang1990" class="citation journal cs1">Zhang, Wei (1990). <a rel="nofollow" class="external text" href="https://drive.google.com/file/d/0B65v6Wo67Tk5ODRzZmhSR29VeDg/view?usp=sharing">"Parallel distributed processing model with local space-invariant interconnections and its optical architecture"</a>. Applied Optics. 29 (32): 4790–7. <a href="/wiki/Bibcode_(identifier)" class="mw-redirect" title="Bibcode (identifier)">Bibcode</a>:<a rel="nofollow" class="external text" href="https://ui.adsabs.harvard.edu/abs/1990ApOpt..29.4790Z">1990ApOpt..29.4790Z</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1364%2FAO.29.004790">10.1364/AO.29.004790</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/20577468">20577468</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20170206111407/https://drive.google.com/file/d/0B65v6Wo67Tk5ODRzZmhSR29VeDg/view?usp=sharing">Archived</a> from the original on 2017-02-06. Retrieved 2016-09-22.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Applied+Optics&rft.atitle=Parallel+distributed+processing+model+with+local+space-invariant+interconnections+and+its+optical+architecture&rft.volume=29&rft.issue=32&rft.pages=4790-7&rft.date=1990&rft_id=info%3Apmid%2F20577468&rft_id=info%3Adoi%2F10.1364%2FAO.29.004790&rft_id=info%3Abibcode%2F1990ApOpt..29.4790Z&rft.aulast=Zhang&rft.aufirst=Wei&rft_id=https%3A%2F%2Fdrive.google.com%2Ffile%2Fd%2F0B65v6Wo67Tk5ODRzZmhSR29VeDg%2Fview%3Fusp%3Dsharing&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-:6-15">^ <a href="#cite_ref-:6_15-0">a</a> <a href="#cite_ref-:6_15-1">b</a> <a href="#cite_ref-:6_15-2">c</a> <a href="#cite_ref-:6_15-3">d</a> <a href="#cite_ref-:6_15-4">e</a> <a href="#cite_ref-:6_15-5">f</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFMoutonMyburghDavel2020" class="citation book cs1">Mouton, Coenraad; Myburgh, Johannes C.; Davel, Marelie H. (2020). <a rel="nofollow" class="external text" href="https://link.springer.com/chapter/10.1007%2F978-3-030-66151-9_17">"Stride and Translation Invariance in CNNs"</a>. In Gerber, Aurona (ed.). Artificial Intelligence Research. Communications in Computer and Information Science. Vol. 1342. Cham: Springer International Publishing. pp. 267–281. <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/2103.10097">2103.10097</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1007%2F978-3-030-66151-9_17">10.1007/978-3-030-66151-9_17</a>. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-3-030-66151-9" title="Special:BookSources/978-3-030-66151-9"><bdi>978-3-030-66151-9</bdi></a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:232269854">232269854</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20210627074505/https://link.springer.com/chapter/10.1007%2F978-3-030-66151-9_17">Archived</a> from the original on 2021-06-27. Retrieved 2021-03-26.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=bookitem&rft.atitle=Stride+and+Translation+Invariance+in+CNNs&rft.btitle=Artificial+Intelligence+Research&rft.place=Cham&rft.series=Communications+in+Computer+and+Information+Science&rft.pages=267-281&rft.pub=Springer+International+Publishing&rft.date=2020&rft_id=info%3Aarxiv%2F2103.10097&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A232269854%23id-name%3DS2CID&rft_id=info%3Adoi%2F10.1007%2F978-3-030-66151-9_17&rft.isbn=978-3-030-66151-9&rft.aulast=Mouton&rft.aufirst=Coenraad&rft.au=Myburgh%2C+Johannes+C.&rft.au=Davel%2C+Marelie+H.&rft_id=https%3A%2F%2Flink.springer.com%2Fchapter%2F10.1007%252F978-3-030-66151-9_17&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-16"><a href="#cite_ref-16">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFKurtzman2019" class="citation journal cs1">Kurtzman, Thomas (August 20, 2019). <a rel="nofollow" class="external text" href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6701836">"Hidden bias in the DUD-E dataset leads to misleading performance of deep learning in structure-based virtual screening"</a>. PLOS ONE. 14 (8): e0220113. <a href="/wiki/Bibcode_(identifier)" class="mw-redirect" title="Bibcode (identifier)">Bibcode</a>:<a rel="nofollow" class="external text" href="https://ui.adsabs.harvard.edu/abs/2019PLoSO..1420113C">2019PLoSO..1420113C</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1371%2Fjournal.pone.0220113">10.1371/journal.pone.0220113</a>. <a href="/wiki/PMC_(identifier)" class="mw-redirect" title="PMC (identifier)">PMC</a> <a rel="nofollow" class="external text" href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6701836">6701836</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/31430292">31430292</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=PLOS+ONE&rft.atitle=Hidden+bias+in+the+DUD-E+dataset+leads+to+misleading+performance+of+deep+learning+in+structure-based+virtual+screening&rft.volume=14&rft.issue=8&rft.pages=e0220113&rft.date=2019-08-20&rft_id=https%3A%2F%2Fwww.ncbi.nlm.nih.gov%2Fpmc%2Farticles%2FPMC6701836%23id-name%3DPMC&rft_id=info%3Apmid%2F31430292&rft_id=info%3Adoi%2F10.1371%2Fjournal.pone.0220113&rft_id=info%3Abibcode%2F2019PLoSO..1420113C&rft.aulast=Kurtzman&rft.aufirst=Thomas&rft_id=https%3A%2F%2Fwww.ncbi.nlm.nih.gov%2Fpmc%2Farticles%2FPMC6701836&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-fukuneoscholar-17">^ <a href="#cite_ref-fukuneoscholar_17-0">a</a> <a href="#cite_ref-fukuneoscholar_17-1">b</a> <a href="#cite_ref-fukuneoscholar_17-2">c</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFFukushima2007" class="citation journal cs1">Fukushima, K. (2007). <a rel="nofollow" class="external text" href="https://doi.org/10.4249%2Fscholarpedia.1717">"Neocognitron"</a>. Scholarpedia. 2 (1): 1717. <a href="/wiki/Bibcode_(identifier)" class="mw-redirect" title="Bibcode (identifier)">Bibcode</a>:<a rel="nofollow" class="external text" href="https://ui.adsabs.harvard.edu/abs/2007SchpJ...2.1717F">2007SchpJ...2.1717F</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.4249%2Fscholarpedia.1717">10.4249/scholarpedia.1717</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Scholarpedia&rft.atitle=Neocognitron&rft.volume=2&rft.issue=1&rft.pages=1717&rft.date=2007&rft_id=info%3Adoi%2F10.4249%2Fscholarpedia.1717&rft_id=info%3Abibcode%2F2007SchpJ...2.1717F&rft.aulast=Fukushima&rft.aufirst=K.&rft_id=https%3A%2F%2Fdoi.org%2F10.4249%252Fscholarpedia.1717&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-hubelwiesel1968-18">^ <a href="#cite_ref-hubelwiesel1968_18-0">a</a> <a href="#cite_ref-hubelwiesel1968_18-1">b</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFHubelWiesel1968" class="citation journal cs1">Hubel, D. H.; Wiesel, T. N. (1968-03-01). <a rel="nofollow" class="external text" href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1557912">"Receptive fields and functional architecture of monkey striate cortex"</a>. The Journal of Physiology. 195 (1): 215–243. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1113%2Fjphysiol.1968.sp008455">10.1113/jphysiol.1968.sp008455</a>. <a href="/wiki/ISSN_(identifier)" class="mw-redirect" title="ISSN (identifier)">ISSN</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/issn/0022-3751">0022-3751</a>. <a href="/wiki/PMC_(identifier)" class="mw-redirect" title="PMC (identifier)">PMC</a> <a rel="nofollow" class="external text" href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1557912">1557912</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/4966457">4966457</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=The+Journal+of+Physiology&rft.atitle=Receptive+fields+and+functional+architecture+of+monkey+striate+cortex&rft.volume=195&rft.issue=1&rft.pages=215-243&rft.date=1968-03-01&rft_id=https%3A%2F%2Fwww.ncbi.nlm.nih.gov%2Fpmc%2Farticles%2FPMC1557912%23id-name%3DPMC&rft.issn=0022-3751&rft_id=info%3Apmid%2F4966457&rft_id=info%3Adoi%2F10.1113%2Fjphysiol.1968.sp008455&rft.aulast=Hubel&rft.aufirst=D.+H.&rft.au=Wiesel%2C+T.+N.&rft_id=https%3A%2F%2Fwww.ncbi.nlm.nih.gov%2Fpmc%2Farticles%2FPMC1557912&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-intro-19">^ <a href="#cite_ref-intro_19-0">a</a> <a href="#cite_ref-intro_19-1">b</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFFukushima1980" class="citation journal cs1">Fukushima, Kunihiko (1980). <a rel="nofollow" class="external text" href="https://www.cs.princeton.edu/courses/archive/spr08/cos598B/Readings/Fukushima1980.pdf">"Neocognitron: A Self-organizing Neural Network Model for a Mechanism of Pattern Recognition Unaffected by Shift in Position"</a> (PDF). Biological Cybernetics. 36 (4): 193–202. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1007%2FBF00344251">10.1007/BF00344251</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/7370364">7370364</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:206775608">206775608</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20140603013137/http://www.cs.princeton.edu/courses/archive/spr08/cos598B/Readings/Fukushima1980.pdf">Archived</a> (PDF) from the original on 3 June 2014. Retrieved 16 November 2013.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Biological+Cybernetics&rft.atitle=Neocognitron%3A+A+Self-organizing+Neural+Network+Model+for+a+Mechanism+of+Pattern+Recognition+Unaffected+by+Shift+in+Position&rft.volume=36&rft.issue=4&rft.pages=193-202&rft.date=1980&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A206775608%23id-name%3DS2CID&rft_id=info%3Apmid%2F7370364&rft_id=info%3Adoi%2F10.1007%2FBF00344251&rft.aulast=Fukushima&rft.aufirst=Kunihiko&rft_id=https%3A%2F%2Fwww.cs.princeton.edu%2Fcourses%2Farchive%2Fspr08%2Fcos598B%2FReadings%2FFukushima1980.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-robust_face_detection-20">^ <a href="#cite_ref-robust_face_detection_20-0">a</a> <a href="#cite_ref-robust_face_detection_20-1">b</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFMatusuguKatsuhiko_MoriYusuke_MitariYuji_Kaneda2003" class="citation journal cs1">Matusugu, Masakazu; Katsuhiko Mori; Yusuke Mitari; Yuji Kaneda (2003). <a rel="nofollow" class="external text" href="http://www.iro.umontreal.ca/~pift6080/H09/documents/papers/sparse/matsugo_etal_face_expression_conv_nnet.pdf">"Subject independent facial expression recognition with robust face detection using a convolutional neural network"</a> (PDF). Neural Networks. 16 (5): 555–559. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1016%2FS0893-6080%2803%2900115-1">10.1016/S0893-6080(03)00115-1</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/12850007">12850007</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20131213022740/http://www.iro.umontreal.ca/~pift6080/H09/documents/papers/sparse/matsugo_etal_face_expression_conv_nnet.pdf">Archived</a> (PDF) from the original on 13 December 2013. Retrieved 17 November 2013.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Neural+Networks&rft.atitle=Subject+independent+facial+expression+recognition+with+robust+face+detection+using+a+convolutional+neural+network&rft.volume=16&rft.issue=5&rft.pages=555-559&rft.date=2003&rft_id=info%3Adoi%2F10.1016%2FS0893-6080%2803%2900115-1&rft_id=info%3Apmid%2F12850007&rft.aulast=Matusugu&rft.aufirst=Masakazu&rft.au=Katsuhiko+Mori&rft.au=Yusuke+Mitari&rft.au=Yuji+Kaneda&rft_id=http%3A%2F%2Fwww.iro.umontreal.ca%2F~pift6080%2FH09%2Fdocuments%2Fpapers%2Fsparse%2Fmatsugo_etal_face_expression_conv_nnet.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-21"><a href="#cite_ref-21">^</a> Convolutional Neural Networks Demystified: A Matched Filtering Perspective Based Tutorial <a rel="nofollow" class="external free" href="https://arxiv.org/abs/2108.11663v3">https://arxiv.org/abs/2108.11663v3</a> </li> <li id="cite_note-deeplearning-22"><a href="#cite_ref-deeplearning_22-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite class="citation web cs1"><a rel="nofollow" class="external text" href="https://web.archive.org/web/20171228091645/http://deeplearning.net/tutorial/lenet.html">"Convolutional Neural Networks (LeNet) – DeepLearning 0.1 documentation"</a>. DeepLearning 0.1. LISA Lab. Archived from <a rel="nofollow" class="external text" href="http://deeplearning.net/tutorial/lenet.html">the original</a> on 28 December 2017. Retrieved 31 August 2013.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=unknown&rft.jtitle=DeepLearning+0.1&rft.atitle=Convolutional+Neural+Networks+%28LeNet%29+%E2%80%93+DeepLearning+0.1+documentation&rft_id=http%3A%2F%2Fdeeplearning.net%2Ftutorial%2Flenet.html&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-23"><a href="#cite_ref-23">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFChollet2017" class="citation arxiv cs1">Chollet, François (2017-04-04). "Xception: Deep Learning with Depthwise Separable Convolutions". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1610.02357">1610.02357</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.CV">cs.CV</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Xception%3A+Deep+Learning+with+Depthwise+Separable+Convolutions&rft.date=2017-04-04&rft_id=info%3Aarxiv%2F1610.02357&rft.aulast=Chollet&rft.aufirst=Fran%C3%A7ois&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-flexible-24">^ <a href="#cite_ref-flexible_24-0">a</a> <a href="#cite_ref-flexible_24-1">b</a> <a href="#cite_ref-flexible_24-2">c</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFCiresanUeli_MeierJonathan_MasciLuca_M._Gambardella2011" class="citation journal cs1">Ciresan, Dan; Ueli Meier; Jonathan Masci; Luca M. Gambardella; Jurgen Schmidhuber (2011). <a rel="nofollow" class="external text" href="https://people.idsia.ch/~juergen/ijcai2011.pdf">"Flexible, High Performance Convolutional Neural Networks for Image Classification"</a> (PDF). Proceedings of the Twenty-Second International Joint Conference on Artificial Intelligence-Volume Volume Two. 2: 1237–1242. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20220405190128/https://people.idsia.ch/~juergen/ijcai2011.pdf">Archived</a> (PDF) from the original on 5 April 2022. Retrieved 17 November 2013.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Proceedings+of+the+Twenty-Second+International+Joint+Conference+on+Artificial+Intelligence-Volume+Volume+Two&rft.atitle=Flexible%2C+High+Performance+Convolutional+Neural+Networks+for+Image+Classification&rft.volume=2&rft.pages=1237-1242&rft.date=2011&rft.aulast=Ciresan&rft.aufirst=Dan&rft.au=Ueli+Meier&rft.au=Jonathan+Masci&rft.au=Luca+M.+Gambardella&rft.au=Jurgen+Schmidhuber&rft_id=https%3A%2F%2Fpeople.idsia.ch%2F~juergen%2Fijcai2011.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-25"><a href="#cite_ref-25">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFKrizhevsky" class="citation web cs1"><a href="/wiki/Alex_Krizhevsky" title="Alex Krizhevsky">Krizhevsky</a>, Alex. <a rel="nofollow" class="external text" href="https://image-net.org/static_files/files/supervision.pdf">"ImageNet Classification with Deep Convolutional Neural Networks"</a> (PDF). <a rel="nofollow" class="external text" href="https://web.archive.org/web/20210425025127/http://www.image-net.org/static_files/files/supervision.pdf">Archived</a> (PDF) from the original on 25 April 2021. Retrieved 17 November 2013.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=unknown&rft.btitle=ImageNet+Classification+with+Deep+Convolutional+Neural+Networks&rft.aulast=Krizhevsky&rft.aufirst=Alex&rft_id=https%3A%2F%2Fimage-net.org%2Fstatic_files%2Ffiles%2Fsupervision.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-Yamaguchi111990-26">^ <a href="#cite_ref-Yamaguchi111990_26-0">a</a> <a href="#cite_ref-Yamaguchi111990_26-1">b</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFYamaguchiSakamotoAkabaneFujimoto1990" class="citation conference cs1">Yamaguchi, Kouichi; Sakamoto, Kenji; Akabane, Toshio; Fujimoto, Yoshiji (November 1990). <a rel="nofollow" class="external text" href="https://web.archive.org/web/20210307233750/https://www.isca-speech.org/archive/icslp_1990/i90_1077.html">A Neural Network for Speaker-Independent Isolated Word Recognition</a>. First International Conference on Spoken Language Processing (ICSLP 90). Kobe, Japan. Archived from <a rel="nofollow" class="external text" href="https://www.isca-speech.org/archive/icslp_1990/i90_1077.html">the original</a> on 2021-03-07. Retrieved 2019-09-04.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=conference&rft.btitle=A+Neural+Network+for+Speaker-Independent+Isolated+Word+Recognition&rft.place=Kobe%2C+Japan&rft.date=1990-11&rft.aulast=Yamaguchi&rft.aufirst=Kouichi&rft.au=Sakamoto%2C+Kenji&rft.au=Akabane%2C+Toshio&rft.au=Fujimoto%2C+Yoshiji&rft_id=https%3A%2F%2Fwww.isca-speech.org%2Farchive%2Ficslp_1990%2Fi90_1077.html&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-mcdns-27">^ <a href="#cite_ref-mcdns_27-0">a</a> <a href="#cite_ref-mcdns_27-1">b</a> <a href="#cite_ref-mcdns_27-2">c</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFCiresanMeierSchmidhuber2012" class="citation book cs1">Ciresan, Dan; Meier, Ueli; Schmidhuber, Jürgen (June 2012). "Multi-column deep neural networks for image classification". 2012 IEEE Conference on Computer Vision and Pattern Recognition. New York, NY: <a href="/wiki/Institute_of_Electrical_and_Electronics_Engineers" title="Institute of Electrical and Electronics Engineers">Institute of Electrical and Electronics Engineers</a> (IEEE). pp. 3642–3649. <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1202.2745">1202.2745</a>. <a href="/wiki/CiteSeerX_(identifier)" class="mw-redirect" title="CiteSeerX (identifier)">CiteSeerX</a> <a rel="nofollow" class="external text" href="https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.300.3283">10.1.1.300.3283</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2FCVPR.2012.6248110">10.1109/CVPR.2012.6248110</a>. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-1-4673-1226-4" title="Special:BookSources/978-1-4673-1226-4"><bdi>978-1-4673-1226-4</bdi></a>. <a href="/wiki/OCLC_(identifier)" class="mw-redirect" title="OCLC (identifier)">OCLC</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/oclc/812295155">812295155</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:2161592">2161592</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=bookitem&rft.atitle=Multi-column+deep+neural+networks+for+image+classification&rft.btitle=2012+IEEE+Conference+on+Computer+Vision+and+Pattern+Recognition&rft.place=New+York%2C+NY&rft.pages=3642-3649&rft.pub=Institute+of+Electrical+and+Electronics+Engineers+%28IEEE%29&rft.date=2012-06&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A2161592%23id-name%3DS2CID&rft_id=info%3Adoi%2F10.1109%2FCVPR.2012.6248110&rft_id=https%3A%2F%2Fciteseerx.ist.psu.edu%2Fviewdoc%2Fsummary%3Fdoi%3D10.1.1.300.3283%23id-name%3DCiteSeerX&rft_id=info%3Aoclcnum%2F812295155&rft_id=info%3Aarxiv%2F1202.2745&rft.isbn=978-1-4673-1226-4&rft.aulast=Ciresan&rft.aufirst=Dan&rft.au=Meier%2C+Ueli&rft.au=Schmidhuber%2C+J%C3%BCrgen&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-28"><a href="#cite_ref-28">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFYuKoltun2016" class="citation arxiv cs1">Yu, Fisher; Koltun, Vladlen (2016-04-30). "Multi-Scale Context Aggregation by Dilated Convolutions". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1511.07122">1511.07122</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.CV">cs.CV</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Multi-Scale+Context+Aggregation+by+Dilated+Convolutions&rft.date=2016-04-30&rft_id=info%3Aarxiv%2F1511.07122&rft.aulast=Yu&rft.aufirst=Fisher&rft.au=Koltun%2C+Vladlen&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-29"><a href="#cite_ref-29">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFChenPapandreouSchroffAdam2017" class="citation arxiv cs1">Chen, Liang-Chieh; Papandreou, George; Schroff, Florian; Adam, Hartwig (2017-12-05). "Rethinking Atrous Convolution for Semantic Image Segmentation". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1706.05587">1706.05587</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.CV">cs.CV</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Rethinking+Atrous+Convolution+for+Semantic+Image+Segmentation&rft.date=2017-12-05&rft_id=info%3Aarxiv%2F1706.05587&rft.aulast=Chen&rft.aufirst=Liang-Chieh&rft.au=Papandreou%2C+George&rft.au=Schroff%2C+Florian&rft.au=Adam%2C+Hartwig&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-30"><a href="#cite_ref-30">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFDutaGeorgescuIonescu2021" class="citation arxiv cs1">Duta, Ionut Cosmin; Georgescu, Mariana Iuliana; Ionescu, Radu Tudor (2021-08-16). "Contextual Convolutional Neural Networks". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/2108.07387">2108.07387</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.CV">cs.CV</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Contextual+Convolutional+Neural+Networks&rft.date=2021-08-16&rft_id=info%3Aarxiv%2F2108.07387&rft.aulast=Duta&rft.aufirst=Ionut+Cosmin&rft.au=Georgescu%2C+Mariana+Iuliana&rft.au=Ionescu%2C+Radu+Tudor&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-LeCun-31"><a href="#cite_ref-LeCun_31-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFLeCun" class="citation web cs1">LeCun, Yann. <a rel="nofollow" class="external text" href="http://yann.lecun.com/exdb/lenet/">"LeNet-5, convolutional neural networks"</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20210224225707/http://yann.lecun.com/exdb/lenet/">Archived</a> from the original on 24 February 2021. Retrieved 16 November 2013.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=unknown&rft.btitle=LeNet-5%2C+convolutional+neural+networks&rft.aulast=LeCun&rft.aufirst=Yann&rft_id=http%3A%2F%2Fyann.lecun.com%2Fexdb%2Flenet%2F&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-32"><a href="#cite_ref-32">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFZeilerTaylorFergus2011" class="citation book cs1">Zeiler, Matthew D.; Taylor, Graham W.; Fergus, Rob (November 2011). <a rel="nofollow" class="external text" href="https://dx.doi.org/10.1109/iccv.2011.6126474">"Adaptive deconvolutional networks for mid and high level feature learning"</a>. 2011 International Conference on Computer Vision. IEEE. pp. 2018–2025. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2Ficcv.2011.6126474">10.1109/iccv.2011.6126474</a>. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-1-4577-1102-2" title="Special:BookSources/978-1-4577-1102-2"><bdi>978-1-4577-1102-2</bdi></a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=bookitem&rft.atitle=Adaptive+deconvolutional+networks+for+mid+and+high+level+feature+learning&rft.btitle=2011+International+Conference+on+Computer+Vision&rft.pages=2018-2025&rft.pub=IEEE&rft.date=2011-11&rft_id=info%3Adoi%2F10.1109%2Ficcv.2011.6126474&rft.isbn=978-1-4577-1102-2&rft.aulast=Zeiler&rft.aufirst=Matthew+D.&rft.au=Taylor%2C+Graham+W.&rft.au=Fergus%2C+Rob&rft_id=http%3A%2F%2Fdx.doi.org%2F10.1109%2Ficcv.2011.6126474&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-33"><a href="#cite_ref-33">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFDumoulinVisin2018" class="citation cs2">Dumoulin, Vincent; Visin, Francesco (2018-01-11), A guide to convolution arithmetic for deep learning, <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1603.07285">1603.07285</a></cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=book&rft.btitle=A+guide+to+convolution+arithmetic+for+deep+learning&rft.date=2018-01-11&rft_id=info%3Aarxiv%2F1603.07285&rft.aulast=Dumoulin&rft.aufirst=Vincent&rft.au=Visin%2C+Francesco&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-34"><a href="#cite_ref-34">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFOdenaDumoulinOlah2016" class="citation journal cs1">Odena, Augustus; Dumoulin, Vincent; Olah, Chris (2016-10-17). <a rel="nofollow" class="external text" href="https://distill.pub/2016/deconv-checkerboard/">"Deconvolution and Checkerboard Artifacts"</a>. Distill. 1 (10): e3. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.23915%2Fdistill.00003">10.23915/distill.00003</a>. <a href="/wiki/ISSN_(identifier)" class="mw-redirect" title="ISSN (identifier)">ISSN</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/issn/2476-0757">2476-0757</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Distill&rft.atitle=Deconvolution+and+Checkerboard+Artifacts&rft.volume=1&rft.issue=10&rft.pages=e3&rft.date=2016-10-17&rft_id=info%3Adoi%2F10.23915%2Fdistill.00003&rft.issn=2476-0757&rft.aulast=Odena&rft.aufirst=Augustus&rft.au=Dumoulin%2C+Vincent&rft.au=Olah%2C+Chris&rft_id=https%3A%2F%2Fdistill.pub%2F2016%2Fdeconv-checkerboard%2F&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-35"><a href="#cite_ref-35">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFvan_DyckKwittDenzlerGruber2021" class="citation journal cs1">van Dyck, Leonard Elia; Kwitt, Roland; Denzler, Sebastian Jochen; Gruber, Walter Roland (2021). <a rel="nofollow" class="external text" href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8526843">"Comparing Object Recognition in Humans and Deep Convolutional Neural Networks—An Eye Tracking Study"</a>. Frontiers in Neuroscience. 15: 750639. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.3389%2Ffnins.2021.750639">10.3389/fnins.2021.750639</a>. <a href="/wiki/ISSN_(identifier)" class="mw-redirect" title="ISSN (identifier)">ISSN</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/issn/1662-453X">1662-453X</a>. <a href="/wiki/PMC_(identifier)" class="mw-redirect" title="PMC (identifier)">PMC</a> <a rel="nofollow" class="external text" href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8526843">8526843</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/34690686">34690686</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Frontiers+in+Neuroscience&rft.atitle=Comparing+Object+Recognition+in+Humans+and+Deep+Convolutional+Neural+Networks%E2%80%94An+Eye+Tracking+Study&rft.volume=15&rft.pages=750639&rft.date=2021&rft_id=https%3A%2F%2Fwww.ncbi.nlm.nih.gov%2Fpmc%2Farticles%2FPMC8526843%23id-name%3DPMC&rft.issn=1662-453X&rft_id=info%3Apmid%2F34690686&rft_id=info%3Adoi%2F10.3389%2Ffnins.2021.750639&rft.aulast=van+Dyck&rft.aufirst=Leonard+Elia&rft.au=Kwitt%2C+Roland&rft.au=Denzler%2C+Sebastian+Jochen&rft.au=Gruber%2C+Walter+Roland&rft_id=https%3A%2F%2Fwww.ncbi.nlm.nih.gov%2Fpmc%2Farticles%2FPMC8526843&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-:4-36">^ <a href="#cite_ref-:4_36-0">a</a> <a href="#cite_ref-:4_36-1">b</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFHubelWiesel1959" class="citation journal cs1">Hubel, DH; Wiesel, TN (October 1959). <a rel="nofollow" class="external text" href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1363130">"Receptive fields of single neurones in the cat's striate cortex"</a>. J. Physiol. 148 (3): 574–91. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1113%2Fjphysiol.1959.sp006308">10.1113/jphysiol.1959.sp006308</a>. <a href="/wiki/PMC_(identifier)" class="mw-redirect" title="PMC (identifier)">PMC</a> <a rel="nofollow" class="external text" href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC1363130">1363130</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/14403679">14403679</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=J.+Physiol.&rft.atitle=Receptive+fields+of+single+neurones+in+the+cat%27s+striate+cortex&rft.volume=148&rft.issue=3&rft.pages=574-91&rft.date=1959-10&rft_id=https%3A%2F%2Fwww.ncbi.nlm.nih.gov%2Fpmc%2Farticles%2FPMC1363130%23id-name%3DPMC&rft_id=info%3Apmid%2F14403679&rft_id=info%3Adoi%2F10.1113%2Fjphysiol.1959.sp006308&rft.aulast=Hubel&rft.aufirst=DH&rft.au=Wiesel%2C+TN&rft_id=https%3A%2F%2Fwww.ncbi.nlm.nih.gov%2Fpmc%2Farticles%2FPMC1363130&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-37"><a href="#cite_ref-37">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFDavid_H._Hubel_and_Torsten_N._Wiesel2005" class="citation book cs1">David H. Hubel and Torsten N. Wiesel (2005). <a rel="nofollow" class="external text" href="https://books.google.com/books?id=8YrxWojxUA4C&pg=PA106">Brain and visual perception: the story of a 25-year collaboration</a>. Oxford University Press US. p. 106. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-0-19-517618-6" title="Special:BookSources/978-0-19-517618-6"><bdi>978-0-19-517618-6</bdi></a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20231016190414/https://books.google.com/books?id=8YrxWojxUA4C&pg=PA106#v=onepage&q&f=false">Archived</a> from the original on 2023-10-16. Retrieved 2019-01-18.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=book&rft.btitle=Brain+and+visual+perception%3A+the+story+of+a+25-year+collaboration&rft.pages=106&rft.pub=Oxford+University+Press+US&rft.date=2005&rft.isbn=978-0-19-517618-6&rft.au=David+H.+Hubel+and+Torsten+N.+Wiesel&rft_id=https%3A%2F%2Fbooks.google.com%2Fbooks%3Fid%3D8YrxWojxUA4C%26pg%3DPA106&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-Fukushima1969-38">^ <a href="#cite_ref-Fukushima1969_38-0">a</a> <a href="#cite_ref-Fukushima1969_38-1">b</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFFukushima1969" class="citation journal cs1">Fukushima, K. (1969). "Visual feature extraction by a multilayered network of analog threshold elements". IEEE Transactions on Systems Science and Cybernetics. 5 (4): 322–333. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2FTSSC.1969.300225">10.1109/TSSC.1969.300225</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=IEEE+Transactions+on+Systems+Science+and+Cybernetics&rft.atitle=Visual+feature+extraction+by+a+multilayered+network+of+analog+threshold+elements&rft.volume=5&rft.issue=4&rft.pages=322-333&rft.date=1969&rft_id=info%3Adoi%2F10.1109%2FTSSC.1969.300225&rft.aulast=Fukushima&rft.aufirst=K.&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-39"><a href="#cite_ref-39">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFRamachandranBarretQuoc2017" class="citation arxiv cs1">Ramachandran, Prajit; Barret, Zoph; Quoc, V. Le (October 16, 2017). "Searching for Activation Functions". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1710.05941">1710.05941</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.NE">cs.NE</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Searching+for+Activation+Functions&rft.date=2017-10-16&rft_id=info%3Aarxiv%2F1710.05941&rft.aulast=Ramachandran&rft.aufirst=Prajit&rft.au=Barret%2C+Zoph&rft.au=Quoc%2C+V.+Le&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-40"><a href="#cite_ref-40">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFFukushima1979" class="citation journal cs1 cs1-prop-long-vol cs1-prop-foreign-lang-source">Fukushima, Kunihiko (October 1979). <a rel="nofollow" class="external text" href="https://search.ieice.org/bin/summary.php?id=j62-a_10_658">"位置ずれに影響されないパターン認識機構の神経回路のモデル --- ネオコグニトロン ---"</a> [Neural network model for a mechanism of pattern recognition unaffected by shift in position — Neocognitron —]. Trans. IECE (in Japanese). J62-A (10): 658–665.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Trans.+IECE&rft.atitle=%E4%BD%8D%E7%BD%AE%E3%81%9A%E3%82%8C%E3%81%AB%E5%BD%B1%E9%9F%BF%E3%81%95%E3%82%8C%E3%81%AA%E3%81%84%E3%83%91%E3%82%BF%E3%83%BC%E3%83%B3%E8%AA%8D%E8%AD%98%E6%A9%9F%E6%A7%8B%E3%81%AE%E7%A5%9E%E7%B5%8C%E5%9B%9E%E8%B7%AF%E3%81%AE%E3%83%A2%E3%83%87%E3%83%AB+---+%E3%83%8D%E3%82%AA%E3%82%B3%E3%82%B0%E3%83%8B%E3%83%88%E3%83%AD%E3%83%B3+---&rft.volume=J62-A&rft.issue=10&rft.pages=658-665&rft.date=1979-10&rft.aulast=Fukushima&rft.aufirst=Kunihiko&rft_id=https%3A%2F%2Fsearch.ieice.org%2Fbin%2Fsummary.php%3Fid%3Dj62-a_10_658&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-weng1993-41"><a href="#cite_ref-weng1993_41-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFWengAhujaHuang1993" class="citation book cs1">Weng, J; Ahuja, N; Huang, TS (1993). <a rel="nofollow" class="external text" href="https://ieeexplore.ieee.org/document/378228">"Learning recognition and segmentation of 3-D objects from 2-D images"</a>. 1993 (4th) International Conference on Computer Vision. IEEE. pp. 121–128. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2FICCV.1993.378228">10.1109/ICCV.1993.378228</a>. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/0-8186-3870-2" title="Special:BookSources/0-8186-3870-2"><bdi>0-8186-3870-2</bdi></a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:8619176">8619176</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=bookitem&rft.atitle=Learning+recognition+and+segmentation+of+3-D+objects+from+2-D+images&rft.btitle=1993+%284th%29+International+Conference+on+Computer+Vision&rft.pages=121-128&rft.pub=IEEE&rft.date=1993&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A8619176%23id-name%3DS2CID&rft_id=info%3Adoi%2F10.1109%2FICCV.1993.378228&rft.isbn=0-8186-3870-2&rft.aulast=Weng&rft.aufirst=J&rft.au=Ahuja%2C+N&rft.au=Huang%2C+TS&rft_id=https%3A%2F%2Fieeexplore.ieee.org%2Fdocument%2F378228&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-schdeepscholar-42">^ <a href="#cite_ref-schdeepscholar_42-0">a</a> <a href="#cite_ref-schdeepscholar_42-1">b</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFSchmidhuber2015" class="citation journal cs1">Schmidhuber, Jürgen (2015). <a rel="nofollow" class="external text" href="http://www.scholarpedia.org/article/Deep_Learning">"Deep Learning"</a>. Scholarpedia. 10 (11): 1527–54. <a href="/wiki/CiteSeerX_(identifier)" class="mw-redirect" title="CiteSeerX (identifier)">CiteSeerX</a> <a rel="nofollow" class="external text" href="https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.76.1541">10.1.1.76.1541</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1162%2Fneco.2006.18.7.1527">10.1162/neco.2006.18.7.1527</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/16764513">16764513</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:2309950">2309950</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20160419024349/http://www.scholarpedia.org/article/Deep_Learning">Archived</a> from the original on 2016-04-19. Retrieved 2019-01-20.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Scholarpedia&rft.atitle=Deep+Learning&rft.volume=10&rft.issue=11&rft.pages=1527-54&rft.date=2015&rft_id=https%3A%2F%2Fciteseerx.ist.psu.edu%2Fviewdoc%2Fsummary%3Fdoi%3D10.1.1.76.1541%23id-name%3DCiteSeerX&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A2309950%23id-name%3DS2CID&rft_id=info%3Apmid%2F16764513&rft_id=info%3Adoi%2F10.1162%2Fneco.2006.18.7.1527&rft.aulast=Schmidhuber&rft.aufirst=J%C3%BCrgen&rft_id=http%3A%2F%2Fwww.scholarpedia.org%2Farticle%2FDeep_Learning&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-Waibel1987-43">^ <a href="#cite_ref-Waibel1987_43-0">a</a> <a href="#cite_ref-Waibel1987_43-1">b</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFWaibel1987" class="citation conference cs1">Waibel, Alex (December 1987). <a rel="nofollow" class="external text" href="https://isl.anthropomatik.kit.edu/pdf/Waibel1987a.pdf">Phoneme Recognition Using Time-Delay Neural Networks</a> (PDF). Meeting of the Institute of Electrical, Information and Communication Engineers (IEICE). Tokyo, Japan.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=conference&rft.btitle=Phoneme+Recognition+Using+Time-Delay+Neural+Networks&rft.place=Tokyo%2C+Japan&rft.date=1987-12&rft.aulast=Waibel&rft.aufirst=Alex&rft_id=https%3A%2F%2Fisl.anthropomatik.kit.edu%2Fpdf%2FWaibel1987a.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-speechsignal-44"><a href="#cite_ref-speechsignal_44-0">^</a> <a href="/wiki/Alex_Waibel" title="Alex Waibel">Alexander Waibel</a> et al., <a rel="nofollow" class="external text" href="http://www.inf.ufrgs.br/~engel/data/media/file/cmp121/waibel89_TDNN.pdf">Phoneme Recognition Using Time-Delay Neural Networks</a> <a rel="nofollow" class="external text" href="https://web.archive.org/web/20210225163001/http://www.inf.ufrgs.br/~engel/data/media/file/cmp121/waibel89_TDNN.pdf">Archived</a> 2021-02-25 at the <a href="/wiki/Wayback_Machine" title="Wayback Machine">Wayback Machine</a> IEEE Transactions on Acoustics, Speech, and Signal Processing, Volume 37, No. 3, pp. 328. - 339 March 1989. </li> <li id="cite_note-45"><a href="#cite_ref-45">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFLeCunBengio1995" class="citation encyclopaedia cs1">LeCun, Yann; Bengio, Yoshua (1995). <a rel="nofollow" class="external text" href="https://www.researchgate.net/publication/2453996">"Convolutional networks for images, speech, and time series"</a>. In Arbib, Michael A. (ed.). The handbook of brain theory and neural networks (Second ed.). The MIT press. pp. 276–278. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20200728164116/https://www.researchgate.net/publication/2453996_Convolutional_Networks_for_Images_Speech_and_Time-Series">Archived</a> from the original on 2020-07-28. Retrieved 2019-12-03.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=bookitem&rft.atitle=Convolutional+networks+for+images%2C+speech%2C+and+time+series&rft.btitle=The+handbook+of+brain+theory+and+neural+networks&rft.pages=276-278&rft.edition=Second&rft.pub=The+MIT+press&rft.date=1995&rft.aulast=LeCun&rft.aufirst=Yann&rft.au=Bengio%2C+Yoshua&rft_id=https%3A%2F%2Fwww.researchgate.net%2Fpublication%2F2453996&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-Hampshire1990-46"><a href="#cite_ref-Hampshire1990_46-0">^</a> John B. Hampshire and Alexander Waibel, <a rel="nofollow" class="external text" href="https://proceedings.neurips.cc/paper/1989/file/979d472a84804b9f647bc185a877a8b5-Paper.pdf">Connectionist Architectures for Multi-Speaker Phoneme Recognition</a> <a rel="nofollow" class="external text" href="https://web.archive.org/web/20220331225059/https://proceedings.neurips.cc/paper/1989/file/979d472a84804b9f647bc185a877a8b5-Paper.pdf">Archived</a> 2022-03-31 at the <a href="/wiki/Wayback_Machine" title="Wayback Machine">Wayback Machine</a>, Advances in Neural Information Processing Systems, 1990, Morgan Kaufmann. </li> <li id="cite_note-Ko2017-47"><a href="#cite_ref-Ko2017_47-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFKoPeddintiPoveySeltzer2018" class="citation conference cs1">Ko, Tom; Peddinti, Vijayaditya; Povey, Daniel; Seltzer, Michael L.; Khudanpur, Sanjeev (March 2018). <a rel="nofollow" class="external text" href="https://www.danielpovey.com/files/2017_icassp_reverberation.pdf">A Study on Data Augmentation of Reverberant Speech for Robust Speech Recognition</a> (PDF). The 42nd IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP 2017). New Orleans, LA, US. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20180708072725/http://danielpovey.com/files/2017_icassp_reverberation.pdf">Archived</a> (PDF) from the original on 2018-07-08. Retrieved 2019-09-04.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=conference&rft.btitle=A+Study+on+Data+Augmentation+of+Reverberant+Speech+for+Robust+Speech+Recognition&rft.place=New+Orleans%2C+LA%2C+US&rft.date=2018-03&rft.aulast=Ko&rft.aufirst=Tom&rft.au=Peddinti%2C+Vijayaditya&rft.au=Povey%2C+Daniel&rft.au=Seltzer%2C+Michael+L.&rft.au=Khudanpur%2C+Sanjeev&rft_id=https%3A%2F%2Fwww.danielpovey.com%2Ffiles%2F2017_icassp_reverberation.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-48"><a href="#cite_ref-48">^</a> Denker, J S, Gardner, W R, Graf, H. P, Henderson, D, Howard, R E, Hubbard, W, Jackel, L D, BaIrd, H S, and Guyon (1989) <a rel="nofollow" class="external text" href="http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.852.5499&rep=rep1&type=pdf">Neural network recognizer for hand-written zip code digits</a> <a rel="nofollow" class="external text" href="https://web.archive.org/web/20180804013916/http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.852.5499&rep=rep1&type=pdf">Archived</a> 2018-08-04 at the <a href="/wiki/Wayback_Machine" title="Wayback Machine">Wayback Machine</a>, AT&T Bell Laboratories </li> <li id="cite_note-:2-49">^ <a href="#cite_ref-:2_49-0">a</a> <a href="#cite_ref-:2_49-1">b</a> Y. LeCun, B. Boser, J. S. Denker, D. Henderson, R. E. Howard, W. Hubbard, L. D. Jackel, <a rel="nofollow" class="external text" href="http://yann.lecun.com/exdb/publis/pdf/lecun-89e.pdf">Backpropagation Applied to Handwritten Zip Code Recognition</a> <a rel="nofollow" class="external text" href="https://web.archive.org/web/20200110090230/http://yann.lecun.com/exdb/publis/pdf/lecun-89e.pdf">Archived</a> 2020-01-10 at the <a href="/wiki/Wayback_Machine" title="Wayback Machine">Wayback Machine</a>; AT&T Bell Laboratories </li> <li id="cite_note-:wz1991-50">^ <a href="#cite_ref-:wz1991_50-0">a</a> <a href="#cite_ref-:wz1991_50-1">b</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFZhang1991" class="citation journal cs1">Zhang, Wei (1991). <a rel="nofollow" class="external text" href="https://drive.google.com/file/d/0B65v6Wo67Tk5cm5DTlNGd0NPUmM/view?usp=sharing">"Image processing of human corneal endothelium based on a learning network"</a>. Applied Optics. 30 (29): 4211–7. <a href="/wiki/Bibcode_(identifier)" class="mw-redirect" title="Bibcode (identifier)">Bibcode</a>:<a rel="nofollow" class="external text" href="https://ui.adsabs.harvard.edu/abs/1991ApOpt..30.4211Z">1991ApOpt..30.4211Z</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1364%2FAO.30.004211">10.1364/AO.30.004211</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/20706526">20706526</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20170206122612/https://drive.google.com/file/d/0B65v6Wo67Tk5cm5DTlNGd0NPUmM/view?usp=sharing">Archived</a> from the original on 2017-02-06. Retrieved 2016-09-22.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Applied+Optics&rft.atitle=Image+processing+of+human+corneal+endothelium+based+on+a+learning+network&rft.volume=30&rft.issue=29&rft.pages=4211-7&rft.date=1991&rft_id=info%3Apmid%2F20706526&rft_id=info%3Adoi%2F10.1364%2FAO.30.004211&rft_id=info%3Abibcode%2F1991ApOpt..30.4211Z&rft.aulast=Zhang&rft.aufirst=Wei&rft_id=https%3A%2F%2Fdrive.google.com%2Ffile%2Fd%2F0B65v6Wo67Tk5cm5DTlNGd0NPUmM%2Fview%3Fusp%3Dsharing&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-:wz1994-51">^ <a href="#cite_ref-:wz1994_51-0">a</a> <a href="#cite_ref-:wz1994_51-1">b</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFZhang1994" class="citation journal cs1">Zhang, Wei (1994). <a rel="nofollow" class="external text" href="https://drive.google.com/file/d/0B65v6Wo67Tk5Ml9qeW5nQ3poVTQ/view?usp=sharing">"Computerized detection of clustered microcalcifications in digital mammograms using a shift-invariant artificial neural network"</a>. Medical Physics. 21 (4): 517–24. <a href="/wiki/Bibcode_(identifier)" class="mw-redirect" title="Bibcode (identifier)">Bibcode</a>:<a rel="nofollow" class="external text" href="https://ui.adsabs.harvard.edu/abs/1994MedPh..21..517Z">1994MedPh..21..517Z</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1118%2F1.597177">10.1118/1.597177</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/8058017">8058017</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20170206030321/https://drive.google.com/file/d/0B65v6Wo67Tk5Ml9qeW5nQ3poVTQ/view?usp=sharing">Archived</a> from the original on 2017-02-06. Retrieved 2016-09-22.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Medical+Physics&rft.atitle=Computerized+detection+of+clustered+microcalcifications+in+digital+mammograms+using+a+shift-invariant+artificial+neural+network&rft.volume=21&rft.issue=4&rft.pages=517-24&rft.date=1994&rft_id=info%3Apmid%2F8058017&rft_id=info%3Adoi%2F10.1118%2F1.597177&rft_id=info%3Abibcode%2F1994MedPh..21..517Z&rft.aulast=Zhang&rft.aufirst=Wei&rft_id=https%3A%2F%2Fdrive.google.com%2Ffile%2Fd%2F0B65v6Wo67Tk5Ml9qeW5nQ3poVTQ%2Fview%3Fusp%3Dsharing&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-lecun95-52">^ <a href="#cite_ref-lecun95_52-0">a</a> <a href="#cite_ref-lecun95_52-1">b</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFLecun,_Y.Jackel,_L._D.Bottou,_L.Cortes,_C.1995" class="citation book cs1">Lecun, Y.; Jackel, L. D.; Bottou, L.; Cortes, C.; Denker, J. S.; Drucker, H.; Guyon, I.; Muller, U. A.; Sackinger, E.; Simard, P.; Vapnik, V. (August 1995). <a rel="nofollow" class="external text" href="http://yann.lecun.com/exdb/publis/pdf/lecun-95a.pdf">Learning algorithms for classification: A comparison on handwritten digit recognition</a> (PDF). World Scientific. pp. 261–276. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1142%2F2808">10.1142/2808</a>. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-981-02-2324-3" title="Special:BookSources/978-981-02-2324-3"><bdi>978-981-02-2324-3</bdi></a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20230502220356/http://yann.lecun.com/exdb/publis/pdf/lecun-95a.pdf">Archived</a> (PDF) from the original on 2 May 2023.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=book&rft.btitle=Learning+algorithms+for+classification%3A+A+comparison+on+handwritten+digit+recognition&rft.pages=261-276&rft.pub=World+Scientific&rft.date=1995-08&rft_id=info%3Adoi%2F10.1142%2F2808&rft.isbn=978-981-02-2324-3&rft.au=Lecun%2C+Y.&rft.au=Jackel%2C+L.+D.&rft.au=Bottou%2C+L.&rft.au=Cortes%2C+C.&rft.au=Denker%2C+J.+S.&rft.au=Drucker%2C+H.&rft.au=Guyon%2C+I.&rft.au=Muller%2C+U.+A.&rft.au=Sackinger%2C+E.&rft.au=Simard%2C+P.&rft.au=Vapnik%2C+V.&rft_id=http%3A%2F%2Fyann.lecun.com%2Fexdb%2Fpublis%2Fpdf%2Flecun-95a.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-53"><a href="#cite_ref-53">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFLecunBottouBengioHaffner1998" class="citation journal cs1">Lecun, Y.; Bottou, L.; Bengio, Y.; Haffner, P. (November 1998). <a rel="nofollow" class="external text" href="https://ieeexplore.ieee.org/document/726791">"Gradient-based learning applied to document recognition"</a>. Proceedings of the IEEE. 86 (11): 2278–2324. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2F5.726791">10.1109/5.726791</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Proceedings+of+the+IEEE&rft.atitle=Gradient-based+learning+applied+to+document+recognition&rft.volume=86&rft.issue=11&rft.pages=2278-2324&rft.date=1998-11&rft_id=info%3Adoi%2F10.1109%2F5.726791&rft.aulast=Lecun&rft.aufirst=Y.&rft.au=Bottou%2C+L.&rft.au=Bengio%2C+Y.&rft.au=Haffner%2C+P.&rft_id=https%3A%2F%2Fieeexplore.ieee.org%2Fdocument%2F726791&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-54"><a href="#cite_ref-54">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFZhang1991" class="citation journal cs1">Zhang, Wei (1991). <a rel="nofollow" class="external text" href="https://drive.google.com/file/d/0B65v6Wo67Tk5dkJTcEMtU2c5Znc/view?usp=sharing">"Error Back Propagation with Minimum-Entropy Weights: A Technique for Better Generalization of 2-D Shift-Invariant NNs"</a>. Proceedings of the International Joint Conference on Neural Networks. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20170206155801/https://drive.google.com/file/d/0B65v6Wo67Tk5dkJTcEMtU2c5Znc/view?usp=sharing">Archived</a> from the original on 2017-02-06. Retrieved 2016-09-22.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Proceedings+of+the+International+Joint+Conference+on+Neural+Networks&rft.atitle=Error+Back+Propagation+with+Minimum-Entropy+Weights%3A+A+Technique+for+Better+Generalization+of+2-D+Shift-Invariant+NNs&rft.date=1991&rft.aulast=Zhang&rft.aufirst=Wei&rft_id=https%3A%2F%2Fdrive.google.com%2Ffile%2Fd%2F0B65v6Wo67Tk5dkJTcEMtU2c5Znc%2Fview%3Fusp%3Dsharing&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-55"><a href="#cite_ref-55">^</a> Daniel Graupe, Ruey Wen Liu, George S Moschytz."<a rel="nofollow" class="external text" href="https://www.researchgate.net/profile/Daniel_Graupe2/publication/241130197_Applications_of_signal_and_image_processing_to_medicine/links/575eef7e08aec91374b42bd2.pdf">Applications of neural networks to medical signal processing</a> <a rel="nofollow" class="external text" href="https://web.archive.org/web/20200728164114/https://www.researchgate.net/profile/Daniel_Graupe2/publication/241130197_Applications_of_signal_and_image_processing_to_medicine/links/575eef7e08aec91374b42bd2.pdf">Archived</a> 2020-07-28 at the <a href="/wiki/Wayback_Machine" title="Wayback Machine">Wayback Machine</a>". In Proc. 27th IEEE Decision and Control Conf., pp. 343–347, 1988. </li> <li id="cite_note-56"><a href="#cite_ref-56">^</a> Daniel Graupe, Boris Vern, G. Gruener, Aaron Field, and Qiu Huang. "<a rel="nofollow" class="external text" href="https://ieeexplore.ieee.org/abstract/document/100522/">Decomposition of surface EMG signals into single fiber action potentials by means of neural network</a> <a rel="nofollow" class="external text" href="https://web.archive.org/web/20190904161656/https://ieeexplore.ieee.org/abstract/document/100522/">Archived</a> 2019-09-04 at the <a href="/wiki/Wayback_Machine" title="Wayback Machine">Wayback Machine</a>". Proc. IEEE International Symp. on Circuits and Systems, pp. 1008–1011, 1989. </li> <li id="cite_note-57"><a href="#cite_ref-57">^</a> Qiu Huang, Daniel Graupe, Yi Fang Huang, Ruey Wen Liu."<a rel="nofollow" class="external text" href="http://www.academia.edu/download/42092095/graupe_huang_q_huang_yf_liu_rw_1989.pdf">Identification of firing patterns of neuronal signals</a>[<a href="/wiki/Wikipedia:Link_rot" title="Wikipedia:Link rot">dead link</a>‍]." In Proc. 28th IEEE Decision and Control Conf., pp. 266–271, 1989. <a rel="nofollow" class="external free" href="https://ieeexplore.ieee.org/document/70115">https://ieeexplore.ieee.org/document/70115</a> <a rel="nofollow" class="external text" href="https://web.archive.org/web/20220331211138/https://ieeexplore.ieee.org/document/70115">Archived</a> 2022-03-31 at the <a href="/wiki/Wayback_Machine" title="Wayback Machine">Wayback Machine</a> </li> <li id="cite_note-58"><a href="#cite_ref-58">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFOhJung2004" class="citation journal cs1">Oh, KS; Jung, K (2004). "GPU implementation of neural networks". Pattern Recognition. 37 (6): 1311–1314. <a href="/wiki/Bibcode_(identifier)" class="mw-redirect" title="Bibcode (identifier)">Bibcode</a>:<a rel="nofollow" class="external text" href="https://ui.adsabs.harvard.edu/abs/2004PatRe..37.1311O">2004PatRe..37.1311O</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1016%2Fj.patcog.2004.01.013">10.1016/j.patcog.2004.01.013</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Pattern+Recognition&rft.atitle=GPU+implementation+of+neural+networks.&rft.volume=37&rft.issue=6&rft.pages=1311-1314&rft.date=2004&rft_id=info%3Adoi%2F10.1016%2Fj.patcog.2004.01.013&rft_id=info%3Abibcode%2F2004PatRe..37.1311O&rft.aulast=Oh&rft.aufirst=KS&rft.au=Jung%2C+K&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-59"><a href="#cite_ref-59">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFDave_SteinkrausPatrice_SimardIan_Buck2005" class="citation conference cs1">Dave Steinkraus; Patrice Simard; Ian Buck (2005). <a rel="nofollow" class="external text" href="https://www.computer.org/csdl/proceedings-article/icdar/2005/24201115/12OmNylKAVX">"Using GPUs for Machine Learning Algorithms"</a>. 12th International Conference on Document Analysis and Recognition (ICDAR 2005). pp. 1115–1119. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2FICDAR.2005.251">10.1109/ICDAR.2005.251</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20220331211138/https://www.computer.org/csdl/proceedings-article/icdar/2005/24201115/12OmNylKAVX">Archived</a> from the original on 2022-03-31. Retrieved 2022-03-31.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=conference&rft.atitle=Using+GPUs+for+Machine+Learning+Algorithms&rft.btitle=12th+International+Conference+on+Document+Analysis+and+Recognition+%28ICDAR+2005%29&rft.pages=1115-1119&rft.date=2005&rft_id=info%3Adoi%2F10.1109%2FICDAR.2005.251&rft.au=Dave+Steinkraus&rft.au=Patrice+Simard&rft.au=Ian+Buck&rft_id=https%3A%2F%2Fwww.computer.org%2Fcsdl%2Fproceedings-article%2Ficdar%2F2005%2F24201115%2F12OmNylKAVX&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-60"><a href="#cite_ref-60">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFKumar_ChellapillaSid_PuriPatrice_Simard2006" class="citation book cs1">Kumar Chellapilla; Sid Puri; Patrice Simard (2006). <a rel="nofollow" class="external text" href="https://hal.inria.fr/inria-00112631/document">"High Performance Convolutional Neural Networks for Document Processing"</a>. In Lorette, Guy (ed.). Tenth International Workshop on Frontiers in Handwriting Recognition. Suvisoft. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20200518193413/https://hal.inria.fr/inria-00112631/document">Archived</a> from the original on 2020-05-18. Retrieved 2016-03-14.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=bookitem&rft.atitle=High+Performance+Convolutional+Neural+Networks+for+Document+Processing&rft.btitle=Tenth+International+Workshop+on+Frontiers+in+Handwriting+Recognition&rft.pub=Suvisoft&rft.date=2006&rft.au=Kumar+Chellapilla&rft.au=Sid+Puri&rft.au=Patrice+Simard&rft_id=https%3A%2F%2Fhal.inria.fr%2Finria-00112631%2Fdocument&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-61"><a href="#cite_ref-61">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFHintonOsinderoTeh2006" class="citation journal cs1">Hinton, GE; Osindero, S; Teh, YW (Jul 2006). "A fast learning algorithm for deep belief nets". Neural Computation. 18 (7): 1527–54. <a href="/wiki/CiteSeerX_(identifier)" class="mw-redirect" title="CiteSeerX (identifier)">CiteSeerX</a> <a rel="nofollow" class="external text" href="https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.76.1541">10.1.1.76.1541</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1162%2Fneco.2006.18.7.1527">10.1162/neco.2006.18.7.1527</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/16764513">16764513</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:2309950">2309950</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Neural+Computation&rft.atitle=A+fast+learning+algorithm+for+deep+belief+nets.&rft.volume=18&rft.issue=7&rft.pages=1527-54&rft.date=2006-07&rft_id=https%3A%2F%2Fciteseerx.ist.psu.edu%2Fviewdoc%2Fsummary%3Fdoi%3D10.1.1.76.1541%23id-name%3DCiteSeerX&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A2309950%23id-name%3DS2CID&rft_id=info%3Apmid%2F16764513&rft_id=info%3Adoi%2F10.1162%2Fneco.2006.18.7.1527&rft.aulast=Hinton&rft.aufirst=GE&rft.au=Osindero%2C+S&rft.au=Teh%2C+YW&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-62"><a href="#cite_ref-62">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFBengioLamblinPopoviciLarochelle2007" class="citation journal cs1">Bengio, Yoshua; Lamblin, Pascal; Popovici, Dan; Larochelle, Hugo (2007). <a rel="nofollow" class="external text" href="https://proceedings.neurips.cc/paper/2006/file/5da713a690c067105aeb2fae32403405-Paper.pdf">"Greedy Layer-Wise Training of Deep Networks"</a> (PDF). Advances in Neural Information Processing Systems: 153–160. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20220602144141/https://proceedings.neurips.cc/paper/2006/file/5da713a690c067105aeb2fae32403405-Paper.pdf">Archived</a> (PDF) from the original on 2022-06-02. Retrieved 2022-03-31.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Advances+in+Neural+Information+Processing+Systems&rft.atitle=Greedy+Layer-Wise+Training+of+Deep+Networks&rft.pages=153-160&rft.date=2007&rft.aulast=Bengio&rft.aufirst=Yoshua&rft.au=Lamblin%2C+Pascal&rft.au=Popovici%2C+Dan&rft.au=Larochelle%2C+Hugo&rft_id=https%3A%2F%2Fproceedings.neurips.cc%2Fpaper%2F2006%2Ffile%2F5da713a690c067105aeb2fae32403405-Paper.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-63"><a href="#cite_ref-63">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFRanzatoPoultneyChopraLeCun2007" class="citation journal cs1">Ranzato, MarcAurelio; Poultney, Christopher; Chopra, Sumit; LeCun, Yann (2007). <a rel="nofollow" class="external text" href="http://yann.lecun.com/exdb/publis/pdf/ranzato-06.pdf">"Efficient Learning of Sparse Representations with an Energy-Based Model"</a> (PDF). Advances in Neural Information Processing Systems. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20160322112400/http://yann.lecun.com/exdb/publis/pdf/ranzato-06.pdf">Archived</a> (PDF) from the original on 2016-03-22. Retrieved 2014-06-26.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Advances+in+Neural+Information+Processing+Systems&rft.atitle=Efficient+Learning+of+Sparse+Representations+with+an+Energy-Based+Model&rft.date=2007&rft.aulast=Ranzato&rft.aufirst=MarcAurelio&rft.au=Poultney%2C+Christopher&rft.au=Chopra%2C+Sumit&rft.au=LeCun%2C+Yann&rft_id=http%3A%2F%2Fyann.lecun.com%2Fexdb%2Fpublis%2Fpdf%2Franzato-06.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-LSD_1-64"><a href="#cite_ref-LSD_1_64-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFRainaMadhavanNg2009" class="citation book cs1">Raina, R; Madhavan, A; Ng, Andrew (14 June 2009). <a rel="nofollow" class="external text" href="http://robotics.stanford.edu/~ang/papers/icml09-LargeScaleUnsupervisedDeepLearningGPU.pdf">"Large-scale deep unsupervised learning using graphics processors"</a> (PDF). Proceedings of the 26th Annual International Conference on Machine Learning. ICML '09: Proceedings of the 26th Annual International Conference on Machine Learning. pp. 873–880. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1145%2F1553374.1553486">10.1145/1553374.1553486</a>. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/9781605585161" title="Special:BookSources/9781605585161"><bdi>9781605585161</bdi></a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:392458">392458</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20201208104513/http://robotics.stanford.edu/~ang/papers/icml09-LargeScaleUnsupervisedDeepLearningGPU.pdf">Archived</a> (PDF) from the original on 8 December 2020. Retrieved 22 December 2023.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=bookitem&rft.atitle=Large-scale+deep+unsupervised+learning+using+graphics+processors&rft.btitle=Proceedings+of+the+26th+Annual+International+Conference+on+Machine+Learning&rft.pages=873-880&rft.pub=ICML+%2709%3A+Proceedings+of+the+26th+Annual+International+Conference+on+Machine+Learning&rft.date=2009-06-14&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A392458%23id-name%3DS2CID&rft_id=info%3Adoi%2F10.1145%2F1553374.1553486&rft.isbn=9781605585161&rft.aulast=Raina&rft.aufirst=R&rft.au=Madhavan%2C+A&rft.au=Ng%2C+Andrew&rft_id=http%3A%2F%2Frobotics.stanford.edu%2F~ang%2Fpapers%2Ficml09-LargeScaleUnsupervisedDeepLearningGPU.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-65"><a href="#cite_ref-65">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFCiresanMeierGambardellaSchmidhuber2010" class="citation journal cs1">Ciresan, Dan; Meier, Ueli; Gambardella, Luca; Schmidhuber, Jürgen (2010). "Deep big simple neural nets for handwritten digit recognition". Neural Computation. 22 (12): 3207–3220. <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1003.0358">1003.0358</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1162%2FNECO_a_00052">10.1162/NECO_a_00052</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/20858131">20858131</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:1918673">1918673</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Neural+Computation&rft.atitle=Deep+big+simple+neural+nets+for+handwritten+digit+recognition.&rft.volume=22&rft.issue=12&rft.pages=3207-3220&rft.date=2010&rft_id=info%3Aarxiv%2F1003.0358&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A1918673%23id-name%3DS2CID&rft_id=info%3Apmid%2F20858131&rft_id=info%3Adoi%2F10.1162%2FNECO_a_00052&rft.aulast=Ciresan&rft.aufirst=Dan&rft.au=Meier%2C+Ueli&rft.au=Gambardella%2C+Luca&rft.au=Schmidhuber%2C+J%C3%BCrgen&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-66"><a href="#cite_ref-66">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite class="citation web cs1"><a rel="nofollow" class="external text" href="https://benchmark.ini.rub.de/gtsrb_results.html">"IJCNN 2011 Competition result table"</a>. OFFICIAL IJCNN2011 COMPETITION. 2010. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20210117024729/https://benchmark.ini.rub.de/gtsrb_results.html">Archived</a> from the original on 2021-01-17. Retrieved 2019-01-14.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=unknown&rft.jtitle=OFFICIAL+IJCNN2011+COMPETITION&rft.atitle=IJCNN+2011+Competition+result+table&rft.date=2010&rft_id=https%3A%2F%2Fbenchmark.ini.rub.de%2Fgtsrb_results.html&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-67"><a href="#cite_ref-67">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFSchmidhuber2017" class="citation web cs1">Schmidhuber, Jürgen (17 March 2017). <a rel="nofollow" class="external text" href="https://people.idsia.ch/~juergen/computer-vision-contests-won-by-gpu-cnns.html">"History of computer vision contests won by deep CNNs on GPU"</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20181219224934/http://people.idsia.ch/~juergen/computer-vision-contests-won-by-gpu-cnns.html">Archived</a> from the original on 19 December 2018. Retrieved 14 January 2019.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=unknown&rft.btitle=History+of+computer+vision+contests+won+by+deep+CNNs+on+GPU&rft.date=2017-03-17&rft.aulast=Schmidhuber&rft.aufirst=J%C3%BCrgen&rft_id=https%3A%2F%2Fpeople.idsia.ch%2F~juergen%2Fcomputer-vision-contests-won-by-gpu-cnns.html&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-:02-68">^ <a href="#cite_ref-:02_68-0">a</a> <a href="#cite_ref-:02_68-1">b</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFKrizhevskySutskeverHinton2017" class="citation journal cs1">Krizhevsky, Alex; Sutskever, Ilya; Hinton, Geoffrey E. (2017-05-24). <a rel="nofollow" class="external text" href="https://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks.pdf">"ImageNet classification with deep convolutional neural networks"</a> (PDF). Communications of the ACM. 60 (6): 84–90. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1145%2F3065386">10.1145/3065386</a>. <a href="/wiki/ISSN_(identifier)" class="mw-redirect" title="ISSN (identifier)">ISSN</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/issn/0001-0782">0001-0782</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:195908774">195908774</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20170516174757/http://papers.nips.cc/paper/4824-imagenet-classification-with-deep-convolutional-neural-networks.pdf">Archived</a> (PDF) from the original on 2017-05-16. Retrieved 2018-12-04.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Communications+of+the+ACM&rft.atitle=ImageNet+classification+with+deep+convolutional+neural+networks&rft.volume=60&rft.issue=6&rft.pages=84-90&rft.date=2017-05-24&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A195908774%23id-name%3DS2CID&rft.issn=0001-0782&rft_id=info%3Adoi%2F10.1145%2F3065386&rft.aulast=Krizhevsky&rft.aufirst=Alex&rft.au=Sutskever%2C+Ilya&rft.au=Hinton%2C+Geoffrey+E.&rft_id=https%3A%2F%2Fpapers.nips.cc%2Fpaper%2F4824-imagenet-classification-with-deep-convolutional-neural-networks.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-69"><a href="#cite_ref-69">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFViebkeMemetiPllanaAbraham2019" class="citation journal cs1">Viebke, Andre; Memeti, Suejb; Pllana, Sabri; Abraham, Ajith (2019). "CHAOS: a parallelization scheme for training convolutional neural networks on Intel Xeon Phi". The Journal of Supercomputing. 75 (1): 197–227. <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1702.07908">1702.07908</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1007%2Fs11227-017-1994-x">10.1007/s11227-017-1994-x</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:14135321">14135321</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=The+Journal+of+Supercomputing&rft.atitle=CHAOS%3A+a+parallelization+scheme+for+training+convolutional+neural+networks+on+Intel+Xeon+Phi&rft.volume=75&rft.issue=1&rft.pages=197-227&rft.date=2019&rft_id=info%3Aarxiv%2F1702.07908&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A14135321%23id-name%3DS2CID&rft_id=info%3Adoi%2F10.1007%2Fs11227-017-1994-x&rft.aulast=Viebke&rft.aufirst=Andre&rft.au=Memeti%2C+Suejb&rft.au=Pllana%2C+Sabri&rft.au=Abraham%2C+Ajith&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-70"><a href="#cite_ref-70">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFViebkePllana2015" class="citation conference cs1">Viebke, Andre; Pllana, Sabri (2015). <a rel="nofollow" class="external text" href="http://lnu.diva-portal.org/smash/record.jsf?pid=diva2%3A877421&dswid=4277">"The Potential of the Intel (R) Xeon Phi for Supervised Deep Learning"</a>. 2015 IEEE 17th International Conference on High Performance Computing and Communications, 2015 IEEE 7th International Symposium on Cyberspace Safety and Security, and 2015 IEEE 12th International Conference on Embedded Software and Systems. IEEE Xplore. IEEE 2015. pp. 758–765. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2FHPCC-CSS-ICESS.2015.45">10.1109/HPCC-CSS-ICESS.2015.45</a>. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-1-4799-8937-9" title="Special:BookSources/978-1-4799-8937-9"><bdi>978-1-4799-8937-9</bdi></a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:15411954">15411954</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20230306003530/http://lnu.diva-portal.org/smash/record.jsf?pid=diva2:877421&dswid=4277">Archived</a> from the original on 2023-03-06. Retrieved 2022-03-31.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=conference&rft.jtitle=IEEE+Xplore&rft.atitle=2015+IEEE+17th+International+Conference+on+High+Performance+Computing+and+Communications%2C+2015+IEEE+7th+International+Symposium+on+Cyberspace+Safety+and+Security%2C+and+2015+IEEE+12th+International+Conference+on+Embedded+Software+and+Systems&rft.pages=758-765&rft.date=2015&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A15411954%23id-name%3DS2CID&rft_id=info%3Adoi%2F10.1109%2FHPCC-CSS-ICESS.2015.45&rft.isbn=978-1-4799-8937-9&rft.aulast=Viebke&rft.aufirst=Andre&rft.au=Pllana%2C+Sabri&rft_id=http%3A%2F%2Flnu.diva-portal.org%2Fsmash%2Frecord.jsf%3Fpid%3Ddiva2%253A877421%26dswid%3D4277&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-71"><a href="#cite_ref-71">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFHinton2012" class="citation journal cs1">Hinton, Geoffrey (2012). <a rel="nofollow" class="external text" href="https://dl.acm.org/doi/10.5555/2999134.2999257">"ImageNet Classification with Deep Convolutional Neural Networks"</a>. NIPS'12: Proceedings of the 25th International Conference on Neural Information Processing Systems - Volume 1. 1: 1097–1105. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20191220014019/https://dl.acm.org/citation.cfm?id=2999134.2999257">Archived</a> from the original on 2019-12-20. Retrieved 2021-03-26 – via ACM.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=NIPS%2712%3A+Proceedings+of+the+25th+International+Conference+on+Neural+Information+Processing+Systems+-+Volume+1&rft.atitle=ImageNet+Classification+with+Deep+Convolutional+Neural+Networks&rft.volume=1&rft.pages=1097-1105&rft.date=2012&rft.aulast=Hinton&rft.aufirst=Geoffrey&rft_id=https%3A%2F%2Fdl.acm.org%2Fdoi%2F10.5555%2F2999134.2999257&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-:5-72">^ <a href="#cite_ref-:5_72-0">a</a> <a href="#cite_ref-:5_72-1">b</a> <a href="#cite_ref-:5_72-2">c</a> <a href="#cite_ref-:5_72-3">d</a> <a href="#cite_ref-:5_72-4">e</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFAzulayWeiss2019" class="citation journal cs1">Azulay, Aharon; Weiss, Yair (2019). <a rel="nofollow" class="external text" href="https://jmlr.org/papers/v20/19-519.html">"Why do deep convolutional networks generalize so poorly to small image transformations?"</a>. Journal of Machine Learning Research. 20 (184): 1–25. <a href="/wiki/ISSN_(identifier)" class="mw-redirect" title="ISSN (identifier)">ISSN</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/issn/1533-7928">1533-7928</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20220331211138/https://jmlr.org/papers/v20/19-519.html">Archived</a> from the original on 2022-03-31. Retrieved 2022-03-31.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Journal+of+Machine+Learning+Research&rft.atitle=Why+do+deep+convolutional+networks+generalize+so+poorly+to+small+image+transformations%3F&rft.volume=20&rft.issue=184&rft.pages=1-25&rft.date=2019&rft.issn=1533-7928&rft.aulast=Azulay&rft.aufirst=Aharon&rft.au=Weiss%2C+Yair&rft_id=https%3A%2F%2Fjmlr.org%2Fpapers%2Fv20%2F19-519.html&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-Géron_Hands-on_ML_2019-73">^ <a href="#cite_ref-Géron_Hands-on_ML_2019_73-0">a</a> <a href="#cite_ref-Géron_Hands-on_ML_2019_73-1">b</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFGéron2019" class="citation book cs1">Géron, Aurélien (2019). Hands-on Machine Learning with Scikit-Learn, Keras, and TensorFlow. Sebastopol, CA: O'Reilly Media. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-1-492-03264-9" title="Special:BookSources/978-1-492-03264-9"><bdi>978-1-492-03264-9</bdi></a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=book&rft.btitle=Hands-on+Machine+Learning+with+Scikit-Learn%2C+Keras%2C+and+TensorFlow&rft.place=Sebastopol%2C+CA&rft.pub=O%27Reilly+Media&rft.date=2019&rft.isbn=978-1-492-03264-9&rft.aulast=G%C3%A9ron&rft.aufirst=Aur%C3%A9lien&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988">, pp. 448 </li> <li id="cite_note-75"><a href="#cite_ref-75">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite class="citation web cs1"><a rel="nofollow" class="external text" href="https://cs231n.github.io/convolutional-networks/">"CS231n Convolutional Neural Networks for Visual Recognition"</a>. cs231n.github.io. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20191023031945/https://cs231n.github.io/convolutional-networks/">Archived</a> from the original on 2019-10-23. Retrieved 2017-04-25.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=unknown&rft.jtitle=cs231n.github.io&rft.atitle=CS231n+Convolutional+Neural+Networks+for+Visual+Recognition&rft_id=https%3A%2F%2Fcs231n.github.io%2Fconvolutional-networks%2F&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-77"><a href="#cite_ref-77">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFNirthikaManivannanRamananWang2022" class="citation journal cs1">Nirthika, Rajendran; Manivannan, Siyamalan; Ramanan, Amirthalingam; Wang, Ruixuan (2022-04-01). <a rel="nofollow" class="external text" href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8804673">"Pooling in convolutional neural networks for medical image analysis: a survey and an empirical study"</a>. Neural Computing and Applications. 34 (7): 5321–5347. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1007%2Fs00521-022-06953-8">10.1007/s00521-022-06953-8</a>. <a href="/wiki/ISSN_(identifier)" class="mw-redirect" title="ISSN (identifier)">ISSN</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/issn/1433-3058">1433-3058</a>. <a href="/wiki/PMC_(identifier)" class="mw-redirect" title="PMC (identifier)">PMC</a> <a rel="nofollow" class="external text" href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8804673">8804673</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/35125669">35125669</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Neural+Computing+and+Applications&rft.atitle=Pooling+in+convolutional+neural+networks+for+medical+image+analysis%3A+a+survey+and+an+empirical+study&rft.volume=34&rft.issue=7&rft.pages=5321-5347&rft.date=2022-04-01&rft_id=https%3A%2F%2Fwww.ncbi.nlm.nih.gov%2Fpmc%2Farticles%2FPMC8804673%23id-name%3DPMC&rft.issn=1433-3058&rft_id=info%3Apmid%2F35125669&rft_id=info%3Adoi%2F10.1007%2Fs00521-022-06953-8&rft.aulast=Nirthika&rft.aufirst=Rajendran&rft.au=Manivannan%2C+Siyamalan&rft.au=Ramanan%2C+Amirthalingam&rft.au=Wang%2C+Ruixuan&rft_id=https%3A%2F%2Fwww.ncbi.nlm.nih.gov%2Fpmc%2Farticles%2FPMC8804673&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-Scherer-ICANN-2010-78">^ <a href="#cite_ref-Scherer-ICANN-2010_78-0">a</a> <a href="#cite_ref-Scherer-ICANN-2010_78-1">b</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFSchererMüllerBehnke2010" class="citation conference cs1">Scherer, Dominik; Müller, Andreas C.; Behnke, Sven (2010). <a rel="nofollow" class="external text" href="http://ais.uni-bonn.de/papers/icann2010_maxpool.pdf">"Evaluation of Pooling Operations in Convolutional Architectures for Object Recognition"</a> (PDF). Artificial Neural Networks (ICANN), 20th International Conference on. Thessaloniki, Greece: Springer. pp. 92–101. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20180403185041/http://ais.uni-bonn.de/papers/icann2010_maxpool.pdf">Archived</a> (PDF) from the original on 2018-04-03. Retrieved 2016-12-28.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=conference&rft.atitle=Evaluation+of+Pooling+Operations+in+Convolutional+Architectures+for+Object+Recognition&rft.btitle=Artificial+Neural+Networks+%28ICANN%29%2C+20th+International+Conference+on&rft.place=Thessaloniki%2C+Greece&rft.pages=92-101&rft.pub=Springer&rft.date=2010&rft.aulast=Scherer&rft.aufirst=Dominik&rft.au=M%C3%BCller%2C+Andreas+C.&rft.au=Behnke%2C+Sven&rft_id=http%3A%2F%2Fais.uni-bonn.de%2Fpapers%2Ficann2010_maxpool.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-79"><a href="#cite_ref-79">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFGraham2014" class="citation arxiv cs1">Graham, Benjamin (2014-12-18). "Fractional Max-Pooling". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1412.6071">1412.6071</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.CV">cs.CV</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Fractional+Max-Pooling&rft.date=2014-12-18&rft_id=info%3Aarxiv%2F1412.6071&rft.aulast=Graham&rft.aufirst=Benjamin&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-80"><a href="#cite_ref-80">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFSpringenbergDosovitskiyBroxRiedmiller2014" class="citation arxiv cs1">Springenberg, Jost Tobias; Dosovitskiy, Alexey; Brox, Thomas; Riedmiller, Martin (2014-12-21). "Striving for Simplicity: The All Convolutional Net". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1412.6806">1412.6806</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.LG">cs.LG</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Striving+for+Simplicity%3A+The+All+Convolutional+Net&rft.date=2014-12-21&rft_id=info%3Aarxiv%2F1412.6806&rft.aulast=Springenberg&rft.aufirst=Jost+Tobias&rft.au=Dosovitskiy%2C+Alexey&rft.au=Brox%2C+Thomas&rft.au=Riedmiller%2C+Martin&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-Ma_Chang_Xie_Ding_2019_pp._3224–3233-81"><a href="#cite_ref-Ma_Chang_Xie_Ding_2019_pp._3224–3233_81-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFMaChangXieDing2019" class="citation journal cs1">Ma, Zhanyu; Chang, Dongliang; Xie, Jiyang; Ding, Yifeng; Wen, Shaoguo; Li, Xiaoxu; Si, Zhongwei; Guo, Jun (2019). "Fine-Grained Vehicle Classification With Channel Max Pooling Modified CNNs". IEEE Transactions on Vehicular Technology. 68 (4). Institute of Electrical and Electronics Engineers (IEEE): 3224–3233. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2Ftvt.2019.2899972">10.1109/tvt.2019.2899972</a>. <a href="/wiki/ISSN_(identifier)" class="mw-redirect" title="ISSN (identifier)">ISSN</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/issn/0018-9545">0018-9545</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:86674074">86674074</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=IEEE+Transactions+on+Vehicular+Technology&rft.atitle=Fine-Grained+Vehicle+Classification+With+Channel+Max+Pooling+Modified+CNNs&rft.volume=68&rft.issue=4&rft.pages=3224-3233&rft.date=2019&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A86674074%23id-name%3DS2CID&rft.issn=0018-9545&rft_id=info%3Adoi%2F10.1109%2Ftvt.2019.2899972&rft.aulast=Ma&rft.aufirst=Zhanyu&rft.au=Chang%2C+Dongliang&rft.au=Xie%2C+Jiyang&rft.au=Ding%2C+Yifeng&rft.au=Wen%2C+Shaoguo&rft.au=Li%2C+Xiaoxu&rft.au=Si%2C+Zhongwei&rft.au=Guo%2C+Jun&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-82"><a href="#cite_ref-82">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFZafarAamirMohd_NawiArshad2022" class="citation journal cs1">Zafar, Afia; Aamir, Muhammad; Mohd Nawi, Nazri; Arshad, Ali; Riaz, Saman; Alruban, Abdulrahman; Dutta, Ashit Kumar; Almotairi, Sultan (2022-08-29). <a rel="nofollow" class="external text" href="https://doi.org/10.3390%2Fapp12178643">"A Comparison of Pooling Methods for Convolutional Neural Networks"</a>. Applied Sciences. 12 (17): 8643. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.3390%2Fapp12178643">10.3390/app12178643</a>. <a href="/wiki/ISSN_(identifier)" class="mw-redirect" title="ISSN (identifier)">ISSN</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/issn/2076-3417">2076-3417</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Applied+Sciences&rft.atitle=A+Comparison+of+Pooling+Methods+for+Convolutional+Neural+Networks&rft.volume=12&rft.issue=17&rft.pages=8643&rft.date=2022-08-29&rft_id=info%3Adoi%2F10.3390%2Fapp12178643&rft.issn=2076-3417&rft.aulast=Zafar&rft.aufirst=Afia&rft.au=Aamir%2C+Muhammad&rft.au=Mohd+Nawi%2C+Nazri&rft.au=Arshad%2C+Ali&rft.au=Riaz%2C+Saman&rft.au=Alruban%2C+Abdulrahman&rft.au=Dutta%2C+Ashit+Kumar&rft.au=Almotairi%2C+Sultan&rft_id=https%3A%2F%2Fdoi.org%2F10.3390%252Fapp12178643&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-83"><a href="#cite_ref-83">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFGholamalinezhadKhosravi2020" class="citation cs2">Gholamalinezhad, Hossein; Khosravi, Hossein (2020-09-16), Pooling Methods in Deep Neural Networks, a Review, <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/2009.07485">2009.07485</a></cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=book&rft.btitle=Pooling+Methods+in+Deep+Neural+Networks%2C+a+Review&rft.date=2020-09-16&rft_id=info%3Aarxiv%2F2009.07485&rft.aulast=Gholamalinezhad&rft.aufirst=Hossein&rft.au=Khosravi%2C+Hossein&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-84"><a href="#cite_ref-84">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFHouseholder1941" class="citation journal cs1">Householder, Alston S. (June 1941). <a rel="nofollow" class="external text" href="http://link.springer.com/10.1007/BF02478220">"A theory of steady-state activity in nerve-fiber networks: I. Definitions and preliminary lemmas"</a>. The Bulletin of Mathematical Biophysics. 3 (2): 63–69. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1007%2FBF02478220">10.1007/BF02478220</a>. <a href="/wiki/ISSN_(identifier)" class="mw-redirect" title="ISSN (identifier)">ISSN</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/issn/0007-4985">0007-4985</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=The+Bulletin+of+Mathematical+Biophysics&rft.atitle=A+theory+of+steady-state+activity+in+nerve-fiber+networks%3A+I.+Definitions+and+preliminary+lemmas&rft.volume=3&rft.issue=2&rft.pages=63-69&rft.date=1941-06&rft_id=info%3Adoi%2F10.1007%2FBF02478220&rft.issn=0007-4985&rft.aulast=Householder&rft.aufirst=Alston+S.&rft_id=http%3A%2F%2Flink.springer.com%2F10.1007%2FBF02478220&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-Romanuke4-85"><a href="#cite_ref-Romanuke4_85-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFRomanuke2017" class="citation journal cs1">Romanuke, Vadim (2017). <a rel="nofollow" class="external text" href="https://doi.org/10.20535%2F1810-0546.2017.1.88156">"Appropriate number and allocation of ReLUs in convolutional neural networks"</a>. Research Bulletin of NTUU "Kyiv Polytechnic Institute". 1 (1): 69–78. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.20535%2F1810-0546.2017.1.88156">10.20535/1810-0546.2017.1.88156</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Research+Bulletin+of+NTUU+%22Kyiv+Polytechnic+Institute%22&rft.atitle=Appropriate+number+and+allocation+of+ReLUs+in+convolutional+neural+networks&rft.volume=1&rft.issue=1&rft.pages=69-78&rft.date=2017&rft_id=info%3Adoi%2F10.20535%2F1810-0546.2017.1.88156&rft.aulast=Romanuke&rft.aufirst=Vadim&rft_id=https%3A%2F%2Fdoi.org%2F10.20535%252F1810-0546.2017.1.88156&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-glorot2011-86"><a href="#cite_ref-glorot2011_86-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFXavier_GlorotAntoine_BordesYoshua_Bengio2011" class="citation conference cs1">Xavier Glorot; Antoine Bordes; <a href="/wiki/Yoshua_Bengio" title="Yoshua Bengio">Yoshua Bengio</a> (2011). <a rel="nofollow" class="external text" href="https://web.archive.org/web/20161213022121/http://jmlr.org/proceedings/papers/v15/glorot11a/glorot11a.pdf">Deep sparse rectifier neural networks</a> (PDF). AISTATS. Archived from <a rel="nofollow" class="external text" href="http://jmlr.org/proceedings/papers/v15/glorot11a/glorot11a.pdf">the original</a> (PDF) on 2016-12-13. Retrieved 2023-04-10. <q>Rectifier and softplus activation functions. The second one is a smooth version of the first.</q></cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=conference&rft.btitle=Deep+sparse+rectifier+neural+networks&rft.date=2011&rft.au=Xavier+Glorot&rft.au=Antoine+Bordes&rft.au=Yoshua+Bengio&rft_id=http%3A%2F%2Fjmlr.org%2Fproceedings%2Fpapers%2Fv15%2Fglorot11a%2Fglorot11a.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-87"><a href="#cite_ref-87">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFKrizhevskySutskever,_I.Hinton,_G._E.2012" class="citation journal cs1">Krizhevsky, A.; Sutskever, I.; Hinton, G. E. (2012). <a rel="nofollow" class="external text" href="https://proceedings.neurips.cc/paper/2012/file/c399862d3b9d6b76c8436e924a68c45b-Paper.pdf">"Imagenet classification with deep convolutional neural networks"</a> (PDF). Advances in Neural Information Processing Systems. 1: 1097–1105. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20220331224736/https://proceedings.neurips.cc/paper/2012/file/c399862d3b9d6b76c8436e924a68c45b-Paper.pdf">Archived</a> (PDF) from the original on 2022-03-31. Retrieved 2022-03-31.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Advances+in+Neural+Information+Processing+Systems&rft.atitle=Imagenet+classification+with+deep+convolutional+neural+networks&rft.volume=1&rft.pages=1097-1105&rft.date=2012&rft.aulast=Krizhevsky&rft.aufirst=A.&rft.au=Sutskever%2C+I.&rft.au=Hinton%2C+G.+E.&rft_id=https%3A%2F%2Fproceedings.neurips.cc%2Fpaper%2F2012%2Ffile%2Fc399862d3b9d6b76c8436e924a68c45b-Paper.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-89"><a href="#cite_ref-89">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFRibeiroSchön2021" class="citation book cs1">Ribeiro, Antonio H.; Schön, Thomas B. (2021). "How Convolutional Neural Networks Deal with Aliasing". ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). pp. 2755–2759. <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/2102.07757">2102.07757</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2FICASSP39728.2021.9414627">10.1109/ICASSP39728.2021.9414627</a>. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-1-7281-7605-5" title="Special:BookSources/978-1-7281-7605-5"><bdi>978-1-7281-7605-5</bdi></a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:231925012">231925012</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=bookitem&rft.atitle=How+Convolutional+Neural+Networks+Deal+with+Aliasing&rft.btitle=ICASSP+2021+-+2021+IEEE+International+Conference+on+Acoustics%2C+Speech+and+Signal+Processing+%28ICASSP%29&rft.pages=2755-2759&rft.date=2021&rft_id=info%3Aarxiv%2F2102.07757&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A231925012%23id-name%3DS2CID&rft_id=info%3Adoi%2F10.1109%2FICASSP39728.2021.9414627&rft.isbn=978-1-7281-7605-5&rft.aulast=Ribeiro&rft.aufirst=Antonio+H.&rft.au=Sch%C3%B6n%2C+Thomas+B.&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-90"><a href="#cite_ref-90">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFMyburghMoutonDavel2020" class="citation book cs1">Myburgh, Johannes C.; Mouton, Coenraad; Davel, Marelie H. (2020). <a rel="nofollow" class="external text" href="https://link.springer.com/chapter/10.1007%2F978-3-030-66151-9_18">"Tracking Translation Invariance in CNNS"</a>. In Gerber, Aurona (ed.). Artificial Intelligence Research. Communications in Computer and Information Science. Vol. 1342. Cham: Springer International Publishing. pp. 282–295. <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/2104.05997">2104.05997</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1007%2F978-3-030-66151-9_18">10.1007/978-3-030-66151-9_18</a>. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-3-030-66151-9" title="Special:BookSources/978-3-030-66151-9"><bdi>978-3-030-66151-9</bdi></a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:233219976">233219976</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20220122015258/http://link.springer.com/chapter/10.1007/978-3-030-66151-9_18">Archived</a> from the original on 2022-01-22. Retrieved 2021-03-26.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=bookitem&rft.atitle=Tracking+Translation+Invariance+in+CNNS&rft.btitle=Artificial+Intelligence+Research&rft.place=Cham&rft.series=Communications+in+Computer+and+Information+Science&rft.pages=282-295&rft.pub=Springer+International+Publishing&rft.date=2020&rft_id=info%3Aarxiv%2F2104.05997&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A233219976%23id-name%3DS2CID&rft_id=info%3Adoi%2F10.1007%2F978-3-030-66151-9_18&rft.isbn=978-3-030-66151-9&rft.aulast=Myburgh&rft.aufirst=Johannes+C.&rft.au=Mouton%2C+Coenraad&rft.au=Davel%2C+Marelie+H.&rft_id=https%3A%2F%2Flink.springer.com%2Fchapter%2F10.1007%252F978-3-030-66151-9_18&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-91"><a href="#cite_ref-91">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFRichard2019" class="citation book cs1">Richard, Zhang (2019-04-25). <a rel="nofollow" class="external text" href="https://www.worldcat.org/oclc/1106340711">Making Convolutional Networks Shift-Invariant Again</a>. <a href="/wiki/OCLC_(identifier)" class="mw-redirect" title="OCLC (identifier)">OCLC</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/oclc/1106340711">1106340711</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=book&rft.btitle=Making+Convolutional+Networks+Shift-Invariant+Again&rft.date=2019-04-25&rft_id=info%3Aoclcnum%2F1106340711&rft.aulast=Richard&rft.aufirst=Zhang&rft_id=https%3A%2F%2Fwww.worldcat.org%2Foclc%2F1106340711&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-92"><a href="#cite_ref-92">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFJadeberg,_Simonyan,_Zisserman,_Kavukcuoglu2015" class="citation journal cs1">Jadeberg, Simonyan, Zisserman, Kavukcuoglu, Max, Karen, Andrew, Koray (2015). <a rel="nofollow" class="external text" href="https://proceedings.neurips.cc/paper/2015/file/33ceb07bf4eeb3da587e268d663aba1a-Paper.pdf">"Spatial Transformer Networks"</a> (PDF). Advances in Neural Information Processing Systems. 28. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20210725115312/https://proceedings.neurips.cc/paper/2015/file/33ceb07bf4eeb3da587e268d663aba1a-Paper.pdf">Archived</a> (PDF) from the original on 2021-07-25. Retrieved 2021-03-26 – via NIPS.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Advances+in+Neural+Information+Processing+Systems&rft.atitle=Spatial+Transformer+Networks&rft.volume=28&rft.date=2015&rft.aulast=Jadeberg%2C+Simonyan%2C+Zisserman%2C+Kavukcuoglu&rft.aufirst=Max%2C+Karen%2C+Andrew%2C+Koray&rft_id=https%3A%2F%2Fproceedings.neurips.cc%2Fpaper%2F2015%2Ffile%2F33ceb07bf4eeb3da587e268d663aba1a-Paper.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"><code class="cs1-code">{{<a href="/wiki/Template:Cite_journal" title="Template:Cite journal">cite journal</a>}}</code>: CS1 maint: multiple names: authors list (<a href="/wiki/Category:CS1_maint:_multiple_names:_authors_list" title="Category:CS1 maint: multiple names: authors list">link</a>) </li> <li id="cite_note-93"><a href="#cite_ref-93">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFE2017" class="citation book cs1">E, Sabour, Sara Frosst, Nicholas Hinton, Geoffrey (2017-10-26). <a rel="nofollow" class="external text" href="https://worldcat.org/oclc/1106278545">Dynamic Routing Between Capsules</a>. <a href="/wiki/OCLC_(identifier)" class="mw-redirect" title="OCLC (identifier)">OCLC</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/oclc/1106278545">1106278545</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=book&rft.btitle=Dynamic+Routing+Between+Capsules&rft.date=2017-10-26&rft_id=info%3Aoclcnum%2F1106278545&rft.aulast=E&rft.aufirst=Sabour%2C+Sara+Frosst%2C+Nicholas+Hinton%2C+Geoffrey&rft_id=https%3A%2F%2Fworldcat.org%2Foclc%2F1106278545&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"><code class="cs1-code">{{<a href="/wiki/Template:Cite_book" title="Template:Cite book">cite book</a>}}</code>: CS1 maint: multiple names: authors list (<a href="/wiki/Category:CS1_maint:_multiple_names:_authors_list" title="Category:CS1 maint: multiple names: authors list">link</a>) </li> <li id="cite_note-94"><a href="#cite_ref-94">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFMatizBarner2019" class="citation journal cs1">Matiz, Sergio; <a href="/wiki/Kenneth_E._Barner" class="mw-redirect" title="Kenneth E. Barner">Barner, Kenneth E.</a> (2019-06-01). <a rel="nofollow" class="external text" href="https://www.sciencedirect.com/science/article/abs/pii/S003132031930055X">"Inductive conformal predictor for convolutional neural networks: Applications to active learning for image classification"</a>. Pattern Recognition. 90: 172–182. <a href="/wiki/Bibcode_(identifier)" class="mw-redirect" title="Bibcode (identifier)">Bibcode</a>:<a rel="nofollow" class="external text" href="https://ui.adsabs.harvard.edu/abs/2019PatRe..90..172M">2019PatRe..90..172M</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1016%2Fj.patcog.2019.01.035">10.1016/j.patcog.2019.01.035</a>. <a href="/wiki/ISSN_(identifier)" class="mw-redirect" title="ISSN (identifier)">ISSN</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/issn/0031-3203">0031-3203</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:127253432">127253432</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20210929092610/https://www.sciencedirect.com/science/article/abs/pii/S003132031930055X">Archived</a> from the original on 2021-09-29. Retrieved 2021-09-29.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Pattern+Recognition&rft.atitle=Inductive+conformal+predictor+for+convolutional+neural+networks%3A+Applications+to+active+learning+for+image+classification&rft.volume=90&rft.pages=172-182&rft.date=2019-06-01&rft_id=info%3Adoi%2F10.1016%2Fj.patcog.2019.01.035&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A127253432%23id-name%3DS2CID&rft.issn=0031-3203&rft_id=info%3Abibcode%2F2019PatRe..90..172M&rft.aulast=Matiz&rft.aufirst=Sergio&rft.au=Barner%2C+Kenneth+E.&rft_id=https%3A%2F%2Fwww.sciencedirect.com%2Fscience%2Farticle%2Fabs%2Fpii%2FS003132031930055X&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-95"><a href="#cite_ref-95">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFWieslanderHarrisonSkogbergJackson2021" class="citation journal cs1">Wieslander, Håkan; Harrison, Philip J.; Skogberg, Gabriel; Jackson, Sonya; Fridén, Markus; Karlsson, Johan; Spjuth, Ola; Wählby, Carolina (February 2021). <a rel="nofollow" class="external text" href="https://doi.org/10.1109%2FJBHI.2020.2996300">"Deep Learning With Conformal Prediction for Hierarchical Analysis of Large-Scale Whole-Slide Tissue Images"</a>. IEEE Journal of Biomedical and Health Informatics. 25 (2): 371–380. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2FJBHI.2020.2996300">10.1109/JBHI.2020.2996300</a>. <a href="/wiki/ISSN_(identifier)" class="mw-redirect" title="ISSN (identifier)">ISSN</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/issn/2168-2208">2168-2208</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/32750907">32750907</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:219885788">219885788</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=IEEE+Journal+of+Biomedical+and+Health+Informatics&rft.atitle=Deep+Learning+With+Conformal+Prediction+for+Hierarchical+Analysis+of+Large-Scale+Whole-Slide+Tissue+Images&rft.volume=25&rft.issue=2&rft.pages=371-380&rft.date=2021-02&rft.issn=2168-2208&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A219885788%23id-name%3DS2CID&rft_id=info%3Apmid%2F32750907&rft_id=info%3Adoi%2F10.1109%2FJBHI.2020.2996300&rft.aulast=Wieslander&rft.aufirst=H%C3%A5kan&rft.au=Harrison%2C+Philip+J.&rft.au=Skogberg%2C+Gabriel&rft.au=Jackson%2C+Sonya&rft.au=Frid%C3%A9n%2C+Markus&rft.au=Karlsson%2C+Johan&rft.au=Spjuth%2C+Ola&rft.au=W%C3%A4hlby%2C+Carolina&rft_id=https%3A%2F%2Fdoi.org%2F10.1109%252FJBHI.2020.2996300&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-96"><a href="#cite_ref-96">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFSrivastavaC._Geoffrey_HintonAlex_KrizhevskyIlya_Sutskever2014" class="citation journal cs1">Srivastava, Nitish; C. Geoffrey Hinton; Alex Krizhevsky; Ilya Sutskever; Ruslan Salakhutdinov (2014). <a rel="nofollow" class="external text" href="http://www.cs.toronto.edu/~rsalakhu/papers/srivastava14a.pdf">"Dropout: A Simple Way to Prevent Neural Networks from overfitting"</a> (PDF). Journal of Machine Learning Research. 15 (1): 1929–1958. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20160119155849/http://www.cs.toronto.edu/~rsalakhu/papers/srivastava14a.pdf">Archived</a> (PDF) from the original on 2016-01-19. Retrieved 2015-01-03.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Journal+of+Machine+Learning+Research&rft.atitle=Dropout%3A+A+Simple+Way+to+Prevent+Neural+Networks+from+overfitting&rft.volume=15&rft.issue=1&rft.pages=1929-1958&rft.date=2014&rft.aulast=Srivastava&rft.aufirst=Nitish&rft.au=C.+Geoffrey+Hinton&rft.au=Alex+Krizhevsky&rft.au=Ilya+Sutskever&rft.au=Ruslan+Salakhutdinov&rft_id=http%3A%2F%2Fwww.cs.toronto.edu%2F~rsalakhu%2Fpapers%2Fsrivastava14a.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-97"><a href="#cite_ref-97">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite class="citation journal cs1"><a rel="nofollow" class="external text" href="http://proceedings.mlr.press/v28/wan13.html">"Regularization of Neural Networks using DropConnect | ICML 2013 | JMLR W&CP"</a>. jmlr.org: 1058–1066. 2013-02-13. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20170812080411/http://proceedings.mlr.press/v28/wan13.html">Archived</a> from the original on 2017-08-12. Retrieved 2015-12-17.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=jmlr.org&rft.atitle=Regularization+of+Neural+Networks+using+DropConnect+%7C+ICML+2013+%7C+JMLR+W%26CP&rft.pages=1058-1066&rft.date=2013-02-13&rft_id=http%3A%2F%2Fproceedings.mlr.press%2Fv28%2Fwan13.html&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-98"><a href="#cite_ref-98">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFZeilerFergus2013" class="citation arxiv cs1">Zeiler, Matthew D.; Fergus, Rob (2013-01-15). "Stochastic Pooling for Regularization of Deep Convolutional Neural Networks". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1301.3557">1301.3557</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.LG">cs.LG</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Stochastic+Pooling+for+Regularization+of+Deep+Convolutional+Neural+Networks&rft.date=2013-01-15&rft_id=info%3Aarxiv%2F1301.3557&rft.aulast=Zeiler&rft.aufirst=Matthew+D.&rft.au=Fergus%2C+Rob&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-:3-99">^ <a href="#cite_ref-:3_99-0">a</a> <a href="#cite_ref-:3_99-1">b</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFPlattSteinkrausSimard2003" class="citation journal cs1">Platt, John; Steinkraus, Dave; Simard, Patrice Y. (August 2003). <a rel="nofollow" class="external text" href="https://www.microsoft.com/en-us/research/publication/best-practices-for-convolutional-neural-networks-applied-to-visual-document-analysis/?from=http%3A%2F%2Fresearch.microsoft.com%2Fapps%2Fpubs%2F%3Fid%3D68920">"Best Practices for Convolutional Neural Networks Applied to Visual Document Analysis – Microsoft Research"</a>. Microsoft Research. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20171107112839/https://www.microsoft.com/en-us/research/publication/best-practices-for-convolutional-neural-networks-applied-to-visual-document-analysis/?from=http%3A%2F%2Fresearch.microsoft.com%2Fapps%2Fpubs%2F%3Fid%3D68920">Archived</a> from the original on 2017-11-07. Retrieved 2015-12-17.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Microsoft+Research&rft.atitle=Best+Practices+for+Convolutional+Neural+Networks+Applied+to+Visual+Document+Analysis+%E2%80%93+Microsoft+Research&rft.date=2003-08&rft.aulast=Platt&rft.aufirst=John&rft.au=Steinkraus%2C+Dave&rft.au=Simard%2C+Patrice+Y.&rft_id=https%3A%2F%2Fwww.microsoft.com%2Fen-us%2Fresearch%2Fpublication%2Fbest-practices-for-convolutional-neural-networks-applied-to-visual-document-analysis%2F%3Ffrom%3Dhttp%253A%252F%252Fresearch.microsoft.com%252Fapps%252Fpubs%252F%253Fid%253D68920&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-100"><a href="#cite_ref-100">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFHintonSrivastavaKrizhevskySutskever2012" class="citation arxiv cs1">Hinton, Geoffrey E.; Srivastava, Nitish; Krizhevsky, Alex; Sutskever, Ilya; Salakhutdinov, Ruslan R. (2012). "Improving neural networks by preventing co-adaptation of feature detectors". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1207.0580">1207.0580</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.NE">cs.NE</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Improving+neural+networks+by+preventing+co-adaptation+of+feature+detectors&rft.date=2012&rft_id=info%3Aarxiv%2F1207.0580&rft.aulast=Hinton&rft.aufirst=Geoffrey+E.&rft.au=Srivastava%2C+Nitish&rft.au=Krizhevsky%2C+Alex&rft.au=Sutskever%2C+Ilya&rft.au=Salakhutdinov%2C+Ruslan+R.&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-101"><a href="#cite_ref-101">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite class="citation web cs1"><a rel="nofollow" class="external text" href="https://jmlr.org/papers/v15/srivastava14a.html">"Dropout: A Simple Way to Prevent Neural Networks from Overfitting"</a>. jmlr.org. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20160305010425/http://jmlr.org/papers/v15/srivastava14a.html">Archived</a> from the original on 2016-03-05. Retrieved 2015-12-17.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=unknown&rft.jtitle=jmlr.org&rft.atitle=Dropout%3A+A+Simple+Way+to+Prevent+Neural+Networks+from+Overfitting&rft_id=https%3A%2F%2Fjmlr.org%2Fpapers%2Fv15%2Fsrivastava14a.html&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-102"><a href="#cite_ref-102">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFHinton1979" class="citation journal cs1">Hinton, Geoffrey (1979). "Some demonstrations of the effects of structural descriptions in mental imagery". Cognitive Science. 3 (3): 231–250. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1016%2Fs0364-0213%2879%2980008-7">10.1016/s0364-0213(79)80008-7</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Cognitive+Science&rft.atitle=Some+demonstrations+of+the+effects+of+structural+descriptions+in+mental+imagery&rft.volume=3&rft.issue=3&rft.pages=231-250&rft.date=1979&rft_id=info%3Adoi%2F10.1016%2Fs0364-0213%2879%2980008-7&rft.aulast=Hinton&rft.aufirst=Geoffrey&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-103"><a href="#cite_ref-103">^</a> Rock, Irvin. "The frame of reference." The legacy of Solomon Asch: Essays in cognition and social psychology (1990): 243–268. </li> <li id="cite_note-104"><a href="#cite_ref-104">^</a> J. Hinton, Coursera lectures on Neural Networks, 2012, Url: <a rel="nofollow" class="external free" href="https://www.coursera.org/learn/neural-networks">https://www.coursera.org/learn/neural-networks</a> <a rel="nofollow" class="external text" href="https://web.archive.org/web/20161231174321/https://www.coursera.org/learn/neural-networks">Archived</a> 2016-12-31 at the <a href="/wiki/Wayback_Machine" title="Wayback Machine">Wayback Machine</a> </li> <li id="cite_note-quartz-105"><a href="#cite_ref-quartz_105-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFDave_Gershgorn2018" class="citation web cs1">Dave Gershgorn (18 June 2018). <a rel="nofollow" class="external text" href="https://qz.com/1307091/the-inside-story-of-how-ai-got-good-enough-to-dominate-silicon-valley/">"The inside story of how AI got good enough to dominate Silicon Valley"</a>. <a href="/wiki/Quartz_(website)" class="mw-redirect" title="Quartz (website)">Quartz</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20191212224842/https://qz.com/1307091/the-inside-story-of-how-ai-got-good-enough-to-dominate-silicon-valley/">Archived</a> from the original on 12 December 2019. Retrieved 5 October 2018.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=unknown&rft.jtitle=Quartz&rft.atitle=The+inside+story+of+how+AI+got+good+enough+to+dominate+Silicon+Valley&rft.date=2018-06-18&rft.au=Dave+Gershgorn&rft_id=https%3A%2F%2Fqz.com%2F1307091%2Fthe-inside-story-of-how-ai-got-good-enough-to-dominate-silicon-valley%2F&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-106"><a href="#cite_ref-106">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFLawrenceC._Lee_GilesAh_Chung_TsoiAndrew_D._Back1997" class="citation journal cs1">Lawrence, Steve; C. Lee Giles; Ah Chung Tsoi; Andrew D. Back (1997). "Face Recognition: A Convolutional Neural Network Approach". IEEE Transactions on Neural Networks. 8 (1): 98–113. <a href="/wiki/CiteSeerX_(identifier)" class="mw-redirect" title="CiteSeerX (identifier)">CiteSeerX</a> <a rel="nofollow" class="external text" href="https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.92.5813">10.1.1.92.5813</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2F72.554195">10.1109/72.554195</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/18255614">18255614</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:2883848">2883848</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=IEEE+Transactions+on+Neural+Networks&rft.atitle=Face+Recognition%3A+A+Convolutional+Neural+Network+Approach&rft.volume=8&rft.issue=1&rft.pages=98-113&rft.date=1997&rft_id=https%3A%2F%2Fciteseerx.ist.psu.edu%2Fviewdoc%2Fsummary%3Fdoi%3D10.1.1.92.5813%23id-name%3DCiteSeerX&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A2883848%23id-name%3DS2CID&rft_id=info%3Apmid%2F18255614&rft_id=info%3Adoi%2F10.1109%2F72.554195&rft.aulast=Lawrence&rft.aufirst=Steve&rft.au=C.+Lee+Giles&rft.au=Ah+Chung+Tsoi&rft.au=Andrew+D.+Back&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-video_quality-107"><a href="#cite_ref-video_quality_107-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFLe_CalletChristian_Viard-GaudinDominique_Barba2006" class="citation journal cs1">Le Callet, Patrick; Christian Viard-Gaudin; Dominique Barba (2006). <a rel="nofollow" class="external text" href="https://hal.archives-ouvertes.fr/file/index/docid/287426/filename/A_convolutional_neural_network_approach_for_objective_video_quality_assessment_completefinal_manuscript.pdf">"A Convolutional Neural Network Approach for Objective Video Quality Assessment"</a> (PDF). IEEE Transactions on Neural Networks. 17 (5): 1316–1327. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2FTNN.2006.879766">10.1109/TNN.2006.879766</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/17001990">17001990</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:221185563">221185563</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20210224123804/https://hal.archives-ouvertes.fr/file/index/docid/287426/filename/A_convolutional_neural_network_approach_for_objective_video_quality_assessment_completefinal_manuscript.pdf">Archived</a> (PDF) from the original on 24 February 2021. Retrieved 17 November 2013.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=IEEE+Transactions+on+Neural+Networks&rft.atitle=A+Convolutional+Neural+Network+Approach+for+Objective+Video+Quality+Assessment&rft.volume=17&rft.issue=5&rft.pages=1316-1327&rft.date=2006&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A221185563%23id-name%3DS2CID&rft_id=info%3Apmid%2F17001990&rft_id=info%3Adoi%2F10.1109%2FTNN.2006.879766&rft.aulast=Le+Callet&rft.aufirst=Patrick&rft.au=Christian+Viard-Gaudin&rft.au=Dominique+Barba&rft_id=https%3A%2F%2Fhal.archives-ouvertes.fr%2Ffile%2Findex%2Fdocid%2F287426%2Ffilename%2FA_convolutional_neural_network_approach_for_objective_video_quality_assessment_completefinal_manuscript.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-ILSVRC2014-108"><a href="#cite_ref-ILSVRC2014_108-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite class="citation web cs1"><a rel="nofollow" class="external text" href="https://image-net.org/challenges/LSVRC/2014/results">"ImageNet Large Scale Visual Recognition Competition 2014 (ILSVRC2014)"</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20160205153105/http://www.image-net.org/challenges/LSVRC/2014/results">Archived</a> from the original on 5 February 2016. Retrieved 30 January 2016.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=unknown&rft.btitle=ImageNet+Large+Scale+Visual+Recognition+Competition+2014+%28ILSVRC2014%29&rft_id=https%3A%2F%2Fimage-net.org%2Fchallenges%2FLSVRC%2F2014%2Fresults&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-googlenet-109"><a href="#cite_ref-googlenet_109-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFSzegedyLiuJiaSermanet2015" class="citation conference cs1">Szegedy, Christian; Liu, Wei; Jia, Yangqing; Sermanet, Pierre; Reed, Scott E.; Anguelov, Dragomir; Erhan, Dumitru; Vanhoucke, Vincent; Rabinovich, Andrew (2015). "Going deeper with convolutions". IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2015, Boston, MA, USA, June 7–12, 2015. IEEE Computer Society. pp. 1–9. <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1409.4842">1409.4842</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2FCVPR.2015.7298594">10.1109/CVPR.2015.7298594</a>. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-1-4673-6964-0" title="Special:BookSources/978-1-4673-6964-0"><bdi>978-1-4673-6964-0</bdi></a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=conference&rft.atitle=Going+deeper+with+convolutions&rft.btitle=IEEE+Conference+on+Computer+Vision+and+Pattern+Recognition%2C+CVPR+2015%2C+Boston%2C+MA%2C+USA%2C+June+7%E2%80%9312%2C+2015&rft.pages=1-9&rft.pub=IEEE+Computer+Society&rft.date=2015&rft_id=info%3Aarxiv%2F1409.4842&rft_id=info%3Adoi%2F10.1109%2FCVPR.2015.7298594&rft.isbn=978-1-4673-6964-0&rft.aulast=Szegedy&rft.aufirst=Christian&rft.au=Liu%2C+Wei&rft.au=Jia%2C+Yangqing&rft.au=Sermanet%2C+Pierre&rft.au=Reed%2C+Scott+E.&rft.au=Anguelov%2C+Dragomir&rft.au=Erhan%2C+Dumitru&rft.au=Vanhoucke%2C+Vincent&rft.au=Rabinovich%2C+Andrew&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-110"><a href="#cite_ref-110">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFRussakovskyDengSuKrause2014" class="citation arxiv cs1"><a href="/wiki/Olga_Russakovsky" title="Olga Russakovsky">Russakovsky, Olga</a>; Deng, Jia; Su, Hao; Krause, Jonathan; Satheesh, Sanjeev; Ma, Sean; Huang, Zhiheng; <a href="/wiki/Andrej_Karpathy" title="Andrej Karpathy">Karpathy, Andrej</a>; Khosla, Aditya; Bernstein, Michael; Berg, Alexander C.; Fei-Fei, Li (2014). "Image Net Large Scale Visual Recognition Challenge". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1409.0575">1409.0575</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.CV">cs.CV</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Image+Net+Large+Scale+Visual+Recognition+Challenge&rft.date=2014&rft_id=info%3Aarxiv%2F1409.0575&rft.aulast=Russakovsky&rft.aufirst=Olga&rft.au=Deng%2C+Jia&rft.au=Su%2C+Hao&rft.au=Krause%2C+Jonathan&rft.au=Satheesh%2C+Sanjeev&rft.au=Ma%2C+Sean&rft.au=Huang%2C+Zhiheng&rft.au=Karpathy%2C+Andrej&rft.au=Khosla%2C+Aditya&rft.au=Bernstein%2C+Michael&rft.au=Berg%2C+Alexander+C.&rft.au=Fei-Fei%2C+Li&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-111"><a href="#cite_ref-111">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite class="citation news cs1"><a rel="nofollow" class="external text" href="https://www.technologyreview.com/2015/02/16/169357/the-face-detection-algorithm-set-to-revolutionize-image-search/">"The Face Detection Algorithm Set To Revolutionize Image Search"</a>. Technology Review. February 16, 2015. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20200920130711/https://www.technologyreview.com/2015/02/16/169357/the-face-detection-algorithm-set-to-revolutionize-image-search/">Archived</a> from the original on 20 September 2020. Retrieved 27 October 2017.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Technology+Review&rft.atitle=The+Face+Detection+Algorithm+Set+To+Revolutionize+Image+Search&rft.date=2015-02-16&rft_id=https%3A%2F%2Fwww.technologyreview.com%2F2015%2F02%2F16%2F169357%2Fthe-face-detection-algorithm-set-to-revolutionize-image-search%2F&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-112"><a href="#cite_ref-112">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFBaccoucheMamaletWolfGarcia2011" class="citation book cs1">Baccouche, Moez; Mamalet, Franck; Wolf, Christian; Garcia, Christophe; Baskurt, Atilla (2011-11-16). "Sequential Deep Learning for Human Action Recognition". In Salah, Albert Ali; Lepri, Bruno (eds.). Human Behavior Unterstanding. Lecture Notes in Computer Science. Vol. 7065. Springer Berlin Heidelberg. pp. 29–39. <a href="/wiki/CiteSeerX_(identifier)" class="mw-redirect" title="CiteSeerX (identifier)">CiteSeerX</a> <a rel="nofollow" class="external text" href="https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.385.4740">10.1.1.385.4740</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1007%2F978-3-642-25446-8_4">10.1007/978-3-642-25446-8_4</a>. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-3-642-25445-1" title="Special:BookSources/978-3-642-25445-1"><bdi>978-3-642-25445-1</bdi></a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=bookitem&rft.atitle=Sequential+Deep+Learning+for+Human+Action+Recognition&rft.btitle=Human+Behavior+Unterstanding&rft.series=Lecture+Notes+in+Computer+Science&rft.pages=29-39&rft.pub=Springer+Berlin+Heidelberg&rft.date=2011-11-16&rft_id=https%3A%2F%2Fciteseerx.ist.psu.edu%2Fviewdoc%2Fsummary%3Fdoi%3D10.1.1.385.4740%23id-name%3DCiteSeerX&rft_id=info%3Adoi%2F10.1007%2F978-3-642-25446-8_4&rft.isbn=978-3-642-25445-1&rft.aulast=Baccouche&rft.aufirst=Moez&rft.au=Mamalet%2C+Franck&rft.au=Wolf%2C+Christian&rft.au=Garcia%2C+Christophe&rft.au=Baskurt%2C+Atilla&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-113"><a href="#cite_ref-113">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFJiXuYangYu2013" class="citation journal cs1">Ji, Shuiwang; Xu, Wei; Yang, Ming; Yu, Kai (2013-01-01). "3D Convolutional Neural Networks for Human Action Recognition". IEEE Transactions on Pattern Analysis and Machine Intelligence. 35 (1): 221–231. <a href="/wiki/CiteSeerX_(identifier)" class="mw-redirect" title="CiteSeerX (identifier)">CiteSeerX</a> <a rel="nofollow" class="external text" href="https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.169.4046">10.1.1.169.4046</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2FTPAMI.2012.59">10.1109/TPAMI.2012.59</a>. <a href="/wiki/ISSN_(identifier)" class="mw-redirect" title="ISSN (identifier)">ISSN</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/issn/0162-8828">0162-8828</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/22392705">22392705</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:1923924">1923924</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=IEEE+Transactions+on+Pattern+Analysis+and+Machine+Intelligence&rft.atitle=3D+Convolutional+Neural+Networks+for+Human+Action+Recognition&rft.volume=35&rft.issue=1&rft.pages=221-231&rft.date=2013-01-01&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A1923924%23id-name%3DS2CID&rft_id=info%3Adoi%2F10.1109%2FTPAMI.2012.59&rft_id=https%3A%2F%2Fciteseerx.ist.psu.edu%2Fviewdoc%2Fsummary%3Fdoi%3D10.1.1.169.4046%23id-name%3DCiteSeerX&rft.issn=0162-8828&rft_id=info%3Apmid%2F22392705&rft.aulast=Ji&rft.aufirst=Shuiwang&rft.au=Xu%2C+Wei&rft.au=Yang%2C+Ming&rft.au=Yu%2C+Kai&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-114"><a href="#cite_ref-114">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFHuangZhouZhangLi2018" class="citation arxiv cs1">Huang, Jie; Zhou, Wengang; Zhang, Qilin; Li, Houqiang; Li, Weiping (2018). "Video-based Sign Language Recognition without Temporal Segmentation". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1801.10111">1801.10111</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.CV">cs.CV</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Video-based+Sign+Language+Recognition+without+Temporal+Segmentation&rft.date=2018&rft_id=info%3Aarxiv%2F1801.10111&rft.aulast=Huang&rft.aufirst=Jie&rft.au=Zhou%2C+Wengang&rft.au=Zhang%2C+Qilin&rft.au=Li%2C+Houqiang&rft.au=Li%2C+Weiping&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-115"><a href="#cite_ref-115">^</a> Karpathy, Andrej, et al. "<a rel="nofollow" class="external text" href="https://www.cv-foundation.org/openaccess/content_cvpr_2014/papers/Karpathy_Large-scale_Video_Classification_2014_CVPR_paper.pdf">Large-scale video classification with convolutional neural networks</a> <a rel="nofollow" class="external text" href="https://web.archive.org/web/20190806022753/https://www.cv-foundation.org/openaccess/content_cvpr_2014/papers/Karpathy_Large-scale_Video_Classification_2014_CVPR_paper.pdf">Archived</a> 2019-08-06 at the <a href="/wiki/Wayback_Machine" title="Wayback Machine">Wayback Machine</a>." IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2014. </li> <li id="cite_note-116"><a href="#cite_ref-116">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFSimonyanZisserman2014" class="citation arxiv cs1">Simonyan, Karen; Zisserman, Andrew (2014). "Two-Stream Convolutional Networks for Action Recognition in Videos". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1406.2199">1406.2199</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.CV">cs.CV</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Two-Stream+Convolutional+Networks+for+Action+Recognition+in+Videos&rft.date=2014&rft_id=info%3Aarxiv%2F1406.2199&rft.aulast=Simonyan&rft.aufirst=Karen&rft.au=Zisserman%2C+Andrew&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> (2014). </li> <li id="cite_note-Wang_Duan_Zhang_Niu_p=1657-117"><a href="#cite_ref-Wang_Duan_Zhang_Niu_p=1657_117-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFWangDuanZhangNiu2018" class="citation journal cs1">Wang, Le; Duan, Xuhuan; Zhang, Qilin; Niu, Zhenxing; Hua, Gang; Zheng, Nanning (2018-05-22). <a rel="nofollow" class="external text" href="https://qilin-zhang.github.io/_pages/pdfs/Segment-Tube_Spatio-Temporal_Action_Localization_in_Untrimmed_Videos_with_Per-Frame_Segmentation.pdf">"Segment-Tube: Spatio-Temporal Action Localization in Untrimmed Videos with Per-Frame Segmentation"</a> (PDF). Sensors. 18 (5): 1657. <a href="/wiki/Bibcode_(identifier)" class="mw-redirect" title="Bibcode (identifier)">Bibcode</a>:<a rel="nofollow" class="external text" href="https://ui.adsabs.harvard.edu/abs/2018Senso..18.1657W">2018Senso..18.1657W</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.3390%2Fs18051657">10.3390/s18051657</a>. <a href="/wiki/ISSN_(identifier)" class="mw-redirect" title="ISSN (identifier)">ISSN</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/issn/1424-8220">1424-8220</a>. <a href="/wiki/PMC_(identifier)" class="mw-redirect" title="PMC (identifier)">PMC</a> <a rel="nofollow" class="external text" href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5982167">5982167</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/29789447">29789447</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20210301195518/https://qilin-zhang.github.io/_pages/pdfs/Segment-Tube_Spatio-Temporal_Action_Localization_in_Untrimmed_Videos_with_Per-Frame_Segmentation.pdf">Archived</a> (PDF) from the original on 2021-03-01. Retrieved 2018-09-14.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Sensors&rft.atitle=Segment-Tube%3A+Spatio-Temporal+Action+Localization+in+Untrimmed+Videos+with+Per-Frame+Segmentation&rft.volume=18&rft.issue=5&rft.pages=1657&rft.date=2018-05-22&rft_id=https%3A%2F%2Fwww.ncbi.nlm.nih.gov%2Fpmc%2Farticles%2FPMC5982167%23id-name%3DPMC&rft_id=info%3Abibcode%2F2018Senso..18.1657W&rft_id=info%3Apmid%2F29789447&rft_id=info%3Adoi%2F10.3390%2Fs18051657&rft.issn=1424-8220&rft.aulast=Wang&rft.aufirst=Le&rft.au=Duan%2C+Xuhuan&rft.au=Zhang%2C+Qilin&rft.au=Niu%2C+Zhenxing&rft.au=Hua%2C+Gang&rft.au=Zheng%2C+Nanning&rft_id=https%3A%2F%2Fqilin-zhang.github.io%2F_pages%2Fpdfs%2FSegment-Tube_Spatio-Temporal_Action_Localization_in_Untrimmed_Videos_with_Per-Frame_Segmentation.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-Duan_Wang_Zhai_Zheng_2018_p.-118"><a href="#cite_ref-Duan_Wang_Zhai_Zheng_2018_p._118-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFDuanWangZhaiZheng2018" class="citation conference cs1">Duan, Xuhuan; Wang, Le; Zhai, Changbo; Zheng, Nanning; Zhang, Qilin; Niu, Zhenxing; Hua, Gang (2018). "Joint Spatio-Temporal Action Localization in Untrimmed Videos with Per-Frame Segmentation". 2018 25th IEEE International Conference on Image Processing (ICIP). 25th IEEE International Conference on Image Processing (ICIP). pp. 918–922. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2Ficip.2018.8451692">10.1109/icip.2018.8451692</a>. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-1-4799-7061-2" title="Special:BookSources/978-1-4799-7061-2"><bdi>978-1-4799-7061-2</bdi></a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=conference&rft.atitle=Joint+Spatio-Temporal+Action+Localization+in+Untrimmed+Videos+with+Per-Frame+Segmentation&rft.btitle=2018+25th+IEEE+International+Conference+on+Image+Processing+%28ICIP%29&rft.pages=918-922&rft.pub=25th+IEEE+International+Conference+on+Image+Processing+%28ICIP%29&rft.date=2018&rft_id=info%3Adoi%2F10.1109%2Ficip.2018.8451692&rft.isbn=978-1-4799-7061-2&rft.aulast=Duan&rft.aufirst=Xuhuan&rft.au=Wang%2C+Le&rft.au=Zhai%2C+Changbo&rft.au=Zheng%2C+Nanning&rft.au=Zhang%2C+Qilin&rft.au=Niu%2C+Zhenxing&rft.au=Hua%2C+Gang&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-119"><a href="#cite_ref-119">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFTaylorFergusLeCunBregler2010" class="citation conference cs1">Taylor, Graham W.; Fergus, Rob; LeCun, Yann; Bregler, Christoph (2010-01-01). <a rel="nofollow" class="external text" href="https://dl.acm.org/doi/10.5555/1888212">Convolutional Learning of Spatio-temporal Features</a>. Proceedings of the 11th European Conference on Computer Vision: Part VI. ECCV'10. Berlin, Heidelberg: Springer-Verlag. pp. 140–153. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-3-642-15566-6" title="Special:BookSources/978-3-642-15566-6"><bdi>978-3-642-15566-6</bdi></a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20220331211137/https://dl.acm.org/doi/10.5555/1888212">Archived</a> from the original on 2022-03-31. Retrieved 2022-03-31.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=conference&rft.btitle=Convolutional+Learning+of+Spatio-temporal+Features&rft.place=Berlin%2C+Heidelberg&rft.series=ECCV%2710&rft.pages=140-153&rft.pub=Springer-Verlag&rft.date=2010-01-01&rft.isbn=978-3-642-15566-6&rft.aulast=Taylor&rft.aufirst=Graham+W.&rft.au=Fergus%2C+Rob&rft.au=LeCun%2C+Yann&rft.au=Bregler%2C+Christoph&rft_id=https%3A%2F%2Fdl.acm.org%2Fdoi%2F10.5555%2F1888212&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-120"><a href="#cite_ref-120">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFLeZouYeungNg2011" class="citation book cs1">Le, Q. V.; Zou, W. Y.; Yeung, S. Y.; Ng, A. Y. (2011-01-01). "Learning hierarchical invariant spatio-temporal features for action recognition with independent subspace analysis". CVPR 2011. CVPR '11. Washington, DC, US: IEEE Computer Society. pp. 3361–3368. <a href="/wiki/CiteSeerX_(identifier)" class="mw-redirect" title="CiteSeerX (identifier)">CiteSeerX</a> <a rel="nofollow" class="external text" href="https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.294.5948">10.1.1.294.5948</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2FCVPR.2011.5995496">10.1109/CVPR.2011.5995496</a>. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-1-4577-0394-2" title="Special:BookSources/978-1-4577-0394-2"><bdi>978-1-4577-0394-2</bdi></a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:6006618">6006618</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=bookitem&rft.atitle=Learning+hierarchical+invariant+spatio-temporal+features+for+action+recognition+with+independent+subspace+analysis&rft.btitle=CVPR+2011&rft.place=Washington%2C+DC%2C+US&rft.series=CVPR+%2711&rft.pages=3361-3368&rft.pub=IEEE+Computer+Society&rft.date=2011-01-01&rft_id=https%3A%2F%2Fciteseerx.ist.psu.edu%2Fviewdoc%2Fsummary%3Fdoi%3D10.1.1.294.5948%23id-name%3DCiteSeerX&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A6006618%23id-name%3DS2CID&rft_id=info%3Adoi%2F10.1109%2FCVPR.2011.5995496&rft.isbn=978-1-4577-0394-2&rft.aulast=Le&rft.aufirst=Q.+V.&rft.au=Zou%2C+W.+Y.&rft.au=Yeung%2C+S.+Y.&rft.au=Ng%2C+A.+Y.&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-121"><a href="#cite_ref-121">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFGrefenstetteBlunsomde_FreitasHermann2014" class="citation arxiv cs1">Grefenstette, Edward; Blunsom, Phil; de Freitas, Nando; Hermann, Karl Moritz (2014-04-29). "A Deep Architecture for Semantic Parsing". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1404.7296">1404.7296</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.CL">cs.CL</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=A+Deep+Architecture+for+Semantic+Parsing&rft.date=2014-04-29&rft_id=info%3Aarxiv%2F1404.7296&rft.aulast=Grefenstette&rft.aufirst=Edward&rft.au=Blunsom%2C+Phil&rft.au=de+Freitas%2C+Nando&rft.au=Hermann%2C+Karl+Moritz&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-122"><a href="#cite_ref-122">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFMesnilDengGaoHe2014" class="citation journal cs1">Mesnil, Gregoire; Deng, Li; Gao, Jianfeng; He, Xiaodong; Shen, Yelong (April 2014). <a rel="nofollow" class="external text" href="https://www.microsoft.com/en-us/research/publication/learning-semantic-representations-using-convolutional-neural-networks-for-web-search/?from=http%3A%2F%2Fresearch.microsoft.com%2Fapps%2Fpubs%2Fdefault.aspx%3Fid%3D214617">"Learning Semantic Representations Using Convolutional Neural Networks for Web Search – Microsoft Research"</a>. Microsoft Research. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20170915160617/https://www.microsoft.com/en-us/research/publication/learning-semantic-representations-using-convolutional-neural-networks-for-web-search/?from=http%3A%2F%2Fresearch.microsoft.com%2Fapps%2Fpubs%2Fdefault.aspx%3Fid%3D214617">Archived</a> from the original on 2017-09-15. Retrieved 2015-12-17.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Microsoft+Research&rft.atitle=Learning+Semantic+Representations+Using+Convolutional+Neural+Networks+for+Web+Search+%E2%80%93+Microsoft+Research&rft.date=2014-04&rft.aulast=Mesnil&rft.aufirst=Gregoire&rft.au=Deng%2C+Li&rft.au=Gao%2C+Jianfeng&rft.au=He%2C+Xiaodong&rft.au=Shen%2C+Yelong&rft_id=https%3A%2F%2Fwww.microsoft.com%2Fen-us%2Fresearch%2Fpublication%2Flearning-semantic-representations-using-convolutional-neural-networks-for-web-search%2F%3Ffrom%3Dhttp%253A%252F%252Fresearch.microsoft.com%252Fapps%252Fpubs%252Fdefault.aspx%253Fid%253D214617&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-123"><a href="#cite_ref-123">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFKalchbrennerGrefenstetteBlunsom2014" class="citation arxiv cs1">Kalchbrenner, Nal; Grefenstette, Edward; Blunsom, Phil (2014-04-08). "A Convolutional Neural Network for Modelling Sentences". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1404.2188">1404.2188</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.CL">cs.CL</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=A+Convolutional+Neural+Network+for+Modelling+Sentences&rft.date=2014-04-08&rft_id=info%3Aarxiv%2F1404.2188&rft.aulast=Kalchbrenner&rft.aufirst=Nal&rft.au=Grefenstette%2C+Edward&rft.au=Blunsom%2C+Phil&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-124"><a href="#cite_ref-124">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFKim2014" class="citation arxiv cs1">Kim, Yoon (2014-08-25). "Convolutional Neural Networks for Sentence Classification". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1408.5882">1408.5882</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.CL">cs.CL</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Convolutional+Neural+Networks+for+Sentence+Classification&rft.date=2014-08-25&rft_id=info%3Aarxiv%2F1408.5882&rft.aulast=Kim&rft.aufirst=Yoon&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-125"><a href="#cite_ref-125">^</a> Collobert, Ronan, and Jason Weston. "<a rel="nofollow" class="external text" href="https://thetalkingmachines.com/sites/default/files/2018-12/unified_nlp.pdf">A unified architecture for natural language processing: Deep neural networks with multitask learning</a> <a rel="nofollow" class="external text" href="https://web.archive.org/web/20190904161653/https://thetalkingmachines.com/sites/default/files/2018-12/unified_nlp.pdf">Archived</a> 2019-09-04 at the <a href="/wiki/Wayback_Machine" title="Wayback Machine">Wayback Machine</a>."Proceedings of the 25th international conference on Machine learning. ACM, 2008. </li> <li id="cite_note-126"><a href="#cite_ref-126">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFCollobertWestonBottouKarlen2011" class="citation arxiv cs1">Collobert, Ronan; Weston, Jason; Bottou, Leon; Karlen, Michael; Kavukcuoglu, Koray; Kuksa, Pavel (2011-03-02). "Natural Language Processing (almost) from Scratch". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1103.0398">1103.0398</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.LG">cs.LG</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Natural+Language+Processing+%28almost%29+from+Scratch&rft.date=2011-03-02&rft_id=info%3Aarxiv%2F1103.0398&rft.aulast=Collobert&rft.aufirst=Ronan&rft.au=Weston%2C+Jason&rft.au=Bottou%2C+Leon&rft.au=Karlen%2C+Michael&rft.au=Kavukcuoglu%2C+Koray&rft.au=Kuksa%2C+Pavel&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-127"><a href="#cite_ref-127">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFYinKannYuSchütze2017" class="citation arxiv cs1">Yin, W; Kann, K; Yu, M; Schütze, H (2017-03-02). "Comparative study of CNN and RNN for natural language processing". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1702.01923">1702.01923</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.LG">cs.LG</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Comparative+study+of+CNN+and+RNN+for+natural+language+processing&rft.date=2017-03-02&rft_id=info%3Aarxiv%2F1702.01923&rft.aulast=Yin&rft.aufirst=W&rft.au=Kann%2C+K&rft.au=Yu%2C+M&rft.au=Sch%C3%BCtze%2C+H&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-128"><a href="#cite_ref-128">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFBaiKolterKoltun2018" class="citation arxiv cs1">Bai, S.; Kolter, J.S.; Koltun, V. (2018). "An empirical evaluation of generic convolutional and recurrent networks for sequence modeling". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1803.01271">1803.01271</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.LG">cs.LG</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=An+empirical+evaluation+of+generic+convolutional+and+recurrent+networks+for+sequence+modeling&rft.date=2018&rft_id=info%3Aarxiv%2F1803.01271&rft.aulast=Bai&rft.aufirst=S.&rft.au=Kolter%2C+J.S.&rft.au=Koltun%2C+V.&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-129"><a href="#cite_ref-129">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFGruber2021" class="citation journal cs1">Gruber, N. (2021). "Detecting dynamics of action in text with a recurrent neural network". Neural Computing and Applications. 33 (12): 15709–15718. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1007%2FS00521-021-06190-5">10.1007/S00521-021-06190-5</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:236307579">236307579</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Neural+Computing+and+Applications&rft.atitle=Detecting+dynamics+of+action+in+text+with+a+recurrent+neural+network&rft.volume=33&rft.issue=12&rft.pages=15709-15718&rft.date=2021&rft_id=info%3Adoi%2F10.1007%2FS00521-021-06190-5&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A236307579%23id-name%3DS2CID&rft.aulast=Gruber&rft.aufirst=N.&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-130"><a href="#cite_ref-130">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFHaotianZhongQianxiao2021" class="citation journal cs1">Haotian, J.; Zhong, Li; Qianxiao, Li (2021). "Approximation Theory of Convolutional Architectures for Time Series Modelling". International Conference on Machine Learning. <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/2107.09355">2107.09355</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=International+Conference+on+Machine+Learning&rft.atitle=Approximation+Theory+of+Convolutional+Architectures+for+Time+Series+Modelling&rft.date=2021&rft_id=info%3Aarxiv%2F2107.09355&rft.aulast=Haotian&rft.aufirst=J.&rft.au=Zhong%2C+Li&rft.au=Qianxiao%2C+Li&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-131"><a href="#cite_ref-131">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFRenXuWangYi2019" class="citation conference cs1">Ren, Hansheng; Xu, Bixiong; Wang, Yujing; Yi, Chao; Huang, Congrui; Kou, Xiaoyu; Xing, Tony; Yang, Mao; Tong, Jie; Zhang, Qi (2019). Time-Series Anomaly Detection Service at Microsoft | Proceedings of the 25th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining. <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1906.03821">1906.03821</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1145%2F3292500.3330680">10.1145/3292500.3330680</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:182952311">182952311</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=conference&rft.btitle=Time-Series+Anomaly+Detection+Service+at+Microsoft+%7C+Proceedings+of+the+25th+ACM+SIGKDD+International+Conference+on+Knowledge+Discovery+%26+Data+Mining&rft.date=2019&rft_id=info%3Aarxiv%2F1906.03821&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A182952311%23id-name%3DS2CID&rft_id=info%3Adoi%2F10.1145%2F3292500.3330680&rft.aulast=Ren&rft.aufirst=Hansheng&rft.au=Xu%2C+Bixiong&rft.au=Wang%2C+Yujing&rft.au=Yi%2C+Chao&rft.au=Huang%2C+Congrui&rft.au=Kou%2C+Xiaoyu&rft.au=Xing%2C+Tony&rft.au=Yang%2C+Mao&rft.au=Tong%2C+Jie&rft.au=Zhang%2C+Qi&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-132"><a href="#cite_ref-132">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFWallachDzambaHeifets2015" class="citation arxiv cs1">Wallach, Izhar; Dzamba, Michael; Heifets, Abraham (2015-10-09). "AtomNet: A Deep Convolutional Neural Network for Bioactivity Prediction in Structure-based Drug Discovery". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1510.02855">1510.02855</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.LG">cs.LG</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=AtomNet%3A+A+Deep+Convolutional+Neural+Network+for+Bioactivity+Prediction+in+Structure-based+Drug+Discovery&rft.date=2015-10-09&rft_id=info%3Aarxiv%2F1510.02855&rft.aulast=Wallach&rft.aufirst=Izhar&rft.au=Dzamba%2C+Michael&rft.au=Heifets%2C+Abraham&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-133"><a href="#cite_ref-133">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFYosinskiCluneNguyenFuchs2015" class="citation arxiv cs1">Yosinski, Jason; Clune, Jeff; Nguyen, Anh; Fuchs, Thomas; Lipson, Hod (2015-06-22). "Understanding Neural Networks Through Deep Visualization". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1506.06579">1506.06579</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.CV">cs.CV</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Understanding+Neural+Networks+Through+Deep+Visualization&rft.date=2015-06-22&rft_id=info%3Aarxiv%2F1506.06579&rft.aulast=Yosinski&rft.aufirst=Jason&rft.au=Clune%2C+Jeff&rft.au=Nguyen%2C+Anh&rft.au=Fuchs%2C+Thomas&rft.au=Lipson%2C+Hod&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-134"><a href="#cite_ref-134">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite class="citation news cs1"><a rel="nofollow" class="external text" href="https://www.theglobeandmail.com/report-on-business/small-business/starting-out/toronto-startup-has-a-faster-way-to-discover-effective-medicines/article25660419/">"Toronto startup has a faster way to discover effective medicines"</a>. The Globe and Mail. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20151020040115/http://www.theglobeandmail.com/report-on-business/small-business/starting-out/toronto-startup-has-a-faster-way-to-discover-effective-medicines/article25660419/">Archived</a> from the original on 2015-10-20. Retrieved 2015-11-09.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=The+Globe+and+Mail&rft.atitle=Toronto+startup+has+a+faster+way+to+discover+effective+medicines&rft_id=https%3A%2F%2Fwww.theglobeandmail.com%2Freport-on-business%2Fsmall-business%2Fstarting-out%2Ftoronto-startup-has-a-faster-way-to-discover-effective-medicines%2Farticle25660419%2F&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-135"><a href="#cite_ref-135">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite class="citation web cs1"><a rel="nofollow" class="external text" href="https://www.kqed.org/futureofyou/3461/startup-harnesses-supercomputers-to-seek-cures">"Startup Harnesses Supercomputers to Seek Cures"</a>. KQED Future of You. 2015-05-27. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20181206234956/https://www.kqed.org/futureofyou/3461/startup-harnesses-supercomputers-to-seek-cures">Archived</a> from the original on 2018-12-06. Retrieved 2015-11-09.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=unknown&rft.jtitle=KQED+Future+of+You&rft.atitle=Startup+Harnesses+Supercomputers+to+Seek+Cures&rft.date=2015-05-27&rft_id=https%3A%2F%2Fwww.kqed.org%2Ffutureofyou%2F3461%2Fstartup-harnesses-supercomputers-to-seek-cures&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-136"><a href="#cite_ref-136">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFChellapillaFogel1999" class="citation journal cs1">Chellapilla, K; Fogel, DB (1999). "Evolving neural networks to play checkers without relying on expert knowledge". IEEE Trans Neural Netw. 10 (6): 1382–91. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2F72.809083">10.1109/72.809083</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/18252639">18252639</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=IEEE+Trans+Neural+Netw&rft.atitle=Evolving+neural+networks+to+play+checkers+without+relying+on+expert+knowledge&rft.volume=10&rft.issue=6&rft.pages=1382-91&rft.date=1999&rft_id=info%3Adoi%2F10.1109%2F72.809083&rft_id=info%3Apmid%2F18252639&rft.aulast=Chellapilla&rft.aufirst=K&rft.au=Fogel%2C+DB&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-137"><a href="#cite_ref-137">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFChellapillaFogel2001" class="citation journal cs1">Chellapilla, K.; Fogel, D.B. (2001). "Evolving an expert checkers playing program without using human expertise". IEEE Transactions on Evolutionary Computation. 5 (4): 422–428. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2F4235.942536">10.1109/4235.942536</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=IEEE+Transactions+on+Evolutionary+Computation&rft.atitle=Evolving+an+expert+checkers+playing+program+without+using+human+expertise&rft.volume=5&rft.issue=4&rft.pages=422-428&rft.date=2001&rft_id=info%3Adoi%2F10.1109%2F4235.942536&rft.aulast=Chellapilla&rft.aufirst=K.&rft.au=Fogel%2C+D.B.&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-138"><a href="#cite_ref-138">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFFogel2001" class="citation book cs1"><a href="/wiki/David_B._Fogel" title="David B. Fogel">Fogel, David</a> (2001). Blondie24: Playing at the Edge of AI. San Francisco, CA: Morgan Kaufmann. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-1558607835" title="Special:BookSources/978-1558607835"><bdi>978-1558607835</bdi></a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=book&rft.btitle=Blondie24%3A+Playing+at+the+Edge+of+AI&rft.place=San+Francisco%2C+CA&rft.pub=Morgan+Kaufmann&rft.date=2001&rft.isbn=978-1558607835&rft.aulast=Fogel&rft.aufirst=David&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-139"><a href="#cite_ref-139">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFClarkStorkey2014" class="citation arxiv cs1">Clark, Christopher; Storkey, Amos (2014). "Teaching Deep Convolutional Neural Networks to Play Go". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1412.3409">1412.3409</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.AI">cs.AI</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Teaching+Deep+Convolutional+Neural+Networks+to+Play+Go&rft.date=2014&rft_id=info%3Aarxiv%2F1412.3409&rft.aulast=Clark&rft.aufirst=Christopher&rft.au=Storkey%2C+Amos&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-140"><a href="#cite_ref-140">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFMaddisonHuangSutskeverSilver2014" class="citation arxiv cs1">Maddison, Chris J.; Huang, Aja; Sutskever, Ilya; Silver, David (2014). "Move Evaluation in Go Using Deep Convolutional Neural Networks". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1412.6564">1412.6564</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.LG">cs.LG</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Move+Evaluation+in+Go+Using+Deep+Convolutional+Neural+Networks&rft.date=2014&rft_id=info%3Aarxiv%2F1412.6564&rft.aulast=Maddison&rft.aufirst=Chris+J.&rft.au=Huang%2C+Aja&rft.au=Sutskever%2C+Ilya&rft.au=Silver%2C+David&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-141"><a href="#cite_ref-141">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite class="citation web cs1"><a rel="nofollow" class="external text" href="https://web.archive.org/web/20160130230207/http://www.deepmind.com/alpha-go.html">"AlphaGo – Google DeepMind"</a>. Archived from <a rel="nofollow" class="external text" href="https://www.deepmind.com/alpha-go.html">the original</a> on 30 January 2016. Retrieved 30 January 2016.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=unknown&rft.btitle=AlphaGo+%E2%80%93+Google+DeepMind&rft_id=https%3A%2F%2Fwww.deepmind.com%2Falpha-go.html&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-142"><a href="#cite_ref-142">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFBaiKolterKoltun2018" class="citation arxiv cs1">Bai, Shaojie; Kolter, J. Zico; Koltun, Vladlen (2018-04-19). "An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1803.01271">1803.01271</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.LG">cs.LG</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=An+Empirical+Evaluation+of+Generic+Convolutional+and+Recurrent+Networks+for+Sequence+Modeling&rft.date=2018-04-19&rft_id=info%3Aarxiv%2F1803.01271&rft.aulast=Bai&rft.aufirst=Shaojie&rft.au=Kolter%2C+J.+Zico&rft.au=Koltun%2C+Vladlen&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-143"><a href="#cite_ref-143">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFYuKoltun2016" class="citation arxiv cs1">Yu, Fisher; Koltun, Vladlen (2016-04-30). "Multi-Scale Context Aggregation by Dilated Convolutions". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1511.07122">1511.07122</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.CV">cs.CV</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Multi-Scale+Context+Aggregation+by+Dilated+Convolutions&rft.date=2016-04-30&rft_id=info%3Aarxiv%2F1511.07122&rft.aulast=Yu&rft.aufirst=Fisher&rft.au=Koltun%2C+Vladlen&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-144"><a href="#cite_ref-144">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFBorovykhBohteOosterlee2018" class="citation arxiv cs1">Borovykh, Anastasia; Bohte, Sander; Oosterlee, Cornelis W. (2018-09-17). "Conditional Time Series Forecasting with Convolutional Neural Networks". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1703.04691">1703.04691</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/stat.ML">stat.ML</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Conditional+Time+Series+Forecasting+with+Convolutional+Neural+Networks&rft.date=2018-09-17&rft_id=info%3Aarxiv%2F1703.04691&rft.aulast=Borovykh&rft.aufirst=Anastasia&rft.au=Bohte%2C+Sander&rft.au=Oosterlee%2C+Cornelis+W.&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-145"><a href="#cite_ref-145">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFMittelman2015" class="citation arxiv cs1">Mittelman, Roni (2015-08-03). "Time-series modeling with undecimated fully convolutional neural networks". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1508.00317">1508.00317</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/stat.ML">stat.ML</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Time-series+modeling+with+undecimated+fully+convolutional+neural+networks&rft.date=2015-08-03&rft_id=info%3Aarxiv%2F1508.00317&rft.aulast=Mittelman&rft.aufirst=Roni&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-146"><a href="#cite_ref-146">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFChenKangChenWang2019" class="citation arxiv cs1">Chen, Yitian; Kang, Yanfei; Chen, Yixiong; Wang, Zizhuo (2019-06-11). "Probabilistic Forecasting with Temporal Convolutional Neural Network". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1906.04397">1906.04397</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/stat.ML">stat.ML</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Probabilistic+Forecasting+with+Temporal+Convolutional+Neural+Network&rft.date=2019-06-11&rft_id=info%3Aarxiv%2F1906.04397&rft.aulast=Chen&rft.aufirst=Yitian&rft.au=Kang%2C+Yanfei&rft.au=Chen%2C+Yixiong&rft.au=Wang%2C+Zizhuo&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-147"><a href="#cite_ref-147">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFZhaoLuChenLiu2017" class="citation journal cs1">Zhao, Bendong; Lu, Huanzhang; Chen, Shangfeng; Liu, Junliang; Wu, Dongya (2017-02-01). "Convolutional neural networks for time series classi". Journal of Systems Engineering and Electronics. 28 (1): 162–169. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.21629%2FJSEE.2017.01.18">10.21629/JSEE.2017.01.18</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Journal+of+Systems+Engineering+and+Electronics&rft.atitle=Convolutional+neural+networks+for+time+series+classi&rft.volume=28&rft.issue=1&rft.pages=162-169&rft.date=2017-02-01&rft_id=info%3Adoi%2F10.21629%2FJSEE.2017.01.18&rft.aulast=Zhao&rft.aufirst=Bendong&rft.au=Lu%2C+Huanzhang&rft.au=Chen%2C+Shangfeng&rft.au=Liu%2C+Junliang&rft.au=Wu%2C+Dongya&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-148"><a href="#cite_ref-148">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFPetneházi2019" class="citation arxiv cs1">Petneházi, Gábor (2019-08-21). "QCNN: Quantile Convolutional Neural Network". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1908.07978">1908.07978</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.LG">cs.LG</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=QCNN%3A+Quantile+Convolutional+Neural+Network&rft.date=2019-08-21&rft_id=info%3Aarxiv%2F1908.07978&rft.aulast=Petneh%C3%A1zi&rft.aufirst=G%C3%A1bor&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-HeiCuBeDa_Hilprecht-149"><a href="#cite_ref-HeiCuBeDa_Hilprecht_149-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFHubert_Mara2019" class="citation cs2 cs1-prop-foreign-lang-source"><a href="/wiki/Hubert_Mara" title="Hubert Mara">Hubert Mara</a> (2019-06-07), HeiCuBeDa Hilprecht – Heidelberg Cuneiform Benchmark Dataset for the Hilprecht Collection (in German), heiDATA – institutional repository for research data of Heidelberg University, <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.11588%2Fdata%2FIE8CCN">10.11588/data/IE8CCN</a></cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=book&rft.btitle=HeiCuBeDa+Hilprecht+%E2%80%93+Heidelberg+Cuneiform+Benchmark+Dataset+for+the+Hilprecht+Collection&rft.pub=heiDATA+%E2%80%93+institutional+repository+for+research+data+of+Heidelberg+University&rft.date=2019-06-07&rft_id=info%3Adoi%2F10.11588%2Fdata%2FIE8CCN&rft.au=Hubert+Mara&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-ICDAR19-150"><a href="#cite_ref-ICDAR19_150-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFHubert_Mara_and_Bartosz_Bogacz2019" class="citation cs2 cs1-prop-foreign-lang-source">Hubert Mara and Bartosz Bogacz (2019), "Breaking the Code on Broken Tablets: The Learning Challenge for Annotated Cuneiform Script in Normalized 2D and 3D Datasets", Proceedings of the 15th International Conference on Document Analysis and Recognition (ICDAR) (in German), Sydney, Australien, pp. 148–153, <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2FICDAR.2019.00032">10.1109/ICDAR.2019.00032</a>, <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-1-7281-3014-9" title="Special:BookSources/978-1-7281-3014-9"><bdi>978-1-7281-3014-9</bdi></a>, <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:211026941">211026941</a></cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Proceedings+of+the+15th+International+Conference+on+Document+Analysis+and+Recognition+%28ICDAR%29&rft.atitle=Breaking+the+Code+on+Broken+Tablets%3A+The+Learning+Challenge+for+Annotated+Cuneiform+Script+in+Normalized+2D+and+3D+Datasets&rft.pages=148-153&rft.date=2019&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A211026941%23id-name%3DS2CID&rft_id=info%3Adoi%2F10.1109%2FICDAR.2019.00032&rft.isbn=978-1-7281-3014-9&rft.au=Hubert+Mara+and+Bartosz+Bogacz&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-ICFHR20-151"><a href="#cite_ref-ICFHR20_151-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFBogaczMara2020" class="citation cs2">Bogacz, Bartosz; Mara, Hubert (2020), "Period Classification of 3D Cuneiform Tablets with Geometric Neural Networks", Proceedings of the 17th International Conference on Frontiers of Handwriting Recognition (ICFHR), Dortmund, Germany</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Proceedings+of+the+17th+International+Conference+on+Frontiers+of+Handwriting+Recognition+%28ICFHR%29&rft.atitle=Period+Classification+of+3D+Cuneiform+Tablets+with+Geometric+Neural+Networks&rft.date=2020&rft.aulast=Bogacz&rft.aufirst=Bartosz&rft.au=Mara%2C+Hubert&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-ICFHR20_Presentation-152"><a href="#cite_ref-ICFHR20_Presentation_152-0">^</a> <a rel="nofollow" class="external text" href="https://www.youtube.com/watch?v=-iFntE51HRw">Presentation of the ICFHR paper on Period Classification of 3D Cuneiform Tablets with Geometric Neural Networks</a> on <a href="/wiki/YouTube_video_(identifier)" class="mw-redirect" title="YouTube video (identifier)">YouTube</a> </li> <li id="cite_note-153"><a href="#cite_ref-153">^</a> Durjoy Sen Maitra; Ujjwal Bhattacharya; S.K. Parui, <a rel="nofollow" class="external text" href="https://ieeexplore.ieee.org/document/7333916">"CNN based common approach to handwritten character recognition of multiple scripts"</a> <a rel="nofollow" class="external text" href="https://web.archive.org/web/20231016190918/https://ieeexplore.ieee.org/document/7333916">Archived</a> 2023-10-16 at the <a href="/wiki/Wayback_Machine" title="Wayback Machine">Wayback Machine</a>, in Document Analysis and Recognition (ICDAR), 2015 13th International Conference on, vol., no., pp.1021–1025, 23–26 Aug. 2015 </li> <li id="cite_note-Interpretable_ML_Symposium_2017-154"><a href="#cite_ref-Interpretable_ML_Symposium_2017_154-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite class="citation web cs1"><a rel="nofollow" class="external text" href="https://web.archive.org/web/20190907063237/http://interpretable.ml/">"NIPS 2017"</a>. Interpretable ML Symposium. 2017-10-20. Archived from <a rel="nofollow" class="external text" href="http://interpretable.ml/">the original</a> on 2019-09-07. Retrieved 2018-09-12.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=unknown&rft.jtitle=Interpretable+ML+Symposium&rft.atitle=NIPS+2017&rft.date=2017-10-20&rft_id=http%3A%2F%2Finterpretable.ml%2F&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-Zang_Wang_Liu_Zhang_2018_pp._97–108-155"><a href="#cite_ref-Zang_Wang_Liu_Zhang_2018_pp._97–108_155-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFZangWangLiuZhang2018" class="citation book cs1">Zang, Jinliang; Wang, Le; Liu, Ziyi; Zhang, Qilin; Hua, Gang; Zheng, Nanning (2018). "Attention-Based Temporal Weighted Convolutional Neural Network for Action Recognition". Artificial Intelligence Applications and Innovations. IFIP Advances in Information and Communication Technology. Vol. 519. Cham: Springer International Publishing. pp. 97–108. <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1803.07179">1803.07179</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1007%2F978-3-319-92007-8_9">10.1007/978-3-319-92007-8_9</a>. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-3-319-92006-1" title="Special:BookSources/978-3-319-92006-1"><bdi>978-3-319-92006-1</bdi></a>. <a href="/wiki/ISSN_(identifier)" class="mw-redirect" title="ISSN (identifier)">ISSN</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/issn/1868-4238">1868-4238</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:4058889">4058889</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=bookitem&rft.atitle=Attention-Based+Temporal+Weighted+Convolutional+Neural+Network+for+Action+Recognition&rft.btitle=Artificial+Intelligence+Applications+and+Innovations&rft.place=Cham&rft.series=IFIP+Advances+in+Information+and+Communication+Technology&rft.pages=97-108&rft.pub=Springer+International+Publishing&rft.date=2018&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A4058889%23id-name%3DS2CID&rft_id=info%3Adoi%2F10.1007%2F978-3-319-92007-8_9&rft_id=info%3Aarxiv%2F1803.07179&rft.issn=1868-4238&rft.isbn=978-3-319-92006-1&rft.aulast=Zang&rft.aufirst=Jinliang&rft.au=Wang%2C+Le&rft.au=Liu%2C+Ziyi&rft.au=Zhang%2C+Qilin&rft.au=Hua%2C+Gang&rft.au=Zheng%2C+Nanning&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-Wang_Zang_Zhang_Niu_p=1979-156"><a href="#cite_ref-Wang_Zang_Zhang_Niu_p=1979_156-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFWangZangZhangNiu2018" class="citation journal cs1">Wang, Le; Zang, Jinliang; Zhang, Qilin; Niu, Zhenxing; Hua, Gang; Zheng, Nanning (2018-06-21). <a rel="nofollow" class="external text" href="https://qilin-zhang.github.io/_pages/pdfs/sensors-18-01979-Action_Recognition_by_an_Attention-Aware_Temporal_Weighted_Convolutional_Neural_Network.pdf">"Action Recognition by an Attention-Aware Temporal Weighted Convolutional Neural Network"</a> (PDF). Sensors. 18 (7): 1979. <a href="/wiki/Bibcode_(identifier)" class="mw-redirect" title="Bibcode (identifier)">Bibcode</a>:<a rel="nofollow" class="external text" href="https://ui.adsabs.harvard.edu/abs/2018Senso..18.1979W">2018Senso..18.1979W</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.3390%2Fs18071979">10.3390/s18071979</a>. <a href="/wiki/ISSN_(identifier)" class="mw-redirect" title="ISSN (identifier)">ISSN</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/issn/1424-8220">1424-8220</a>. <a href="/wiki/PMC_(identifier)" class="mw-redirect" title="PMC (identifier)">PMC</a> <a rel="nofollow" class="external text" href="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6069475">6069475</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/29933555">29933555</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20180913040055/https://qilin-zhang.github.io/_pages/pdfs/sensors-18-01979-Action_Recognition_by_an_Attention-Aware_Temporal_Weighted_Convolutional_Neural_Network.pdf">Archived</a> (PDF) from the original on 2018-09-13. Retrieved 2018-09-14.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Sensors&rft.atitle=Action+Recognition+by+an+Attention-Aware+Temporal+Weighted+Convolutional+Neural+Network&rft.volume=18&rft.issue=7&rft.pages=1979&rft.date=2018-06-21&rft_id=https%3A%2F%2Fwww.ncbi.nlm.nih.gov%2Fpmc%2Farticles%2FPMC6069475%23id-name%3DPMC&rft_id=info%3Abibcode%2F2018Senso..18.1979W&rft_id=info%3Apmid%2F29933555&rft_id=info%3Adoi%2F10.3390%2Fs18071979&rft.issn=1424-8220&rft.aulast=Wang&rft.aufirst=Le&rft.au=Zang%2C+Jinliang&rft.au=Zhang%2C+Qilin&rft.au=Niu%2C+Zhenxing&rft.au=Hua%2C+Gang&rft.au=Zheng%2C+Nanning&rft_id=https%3A%2F%2Fqilin-zhang.github.io%2F_pages%2Fpdfs%2Fsensors-18-01979-Action_Recognition_by_an_Attention-Aware_Temporal_Weighted_Convolutional_Neural_Network.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-Ong_Chavez_Hong_2015-157"><a href="#cite_ref-Ong_Chavez_Hong_2015_157-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFOngChavezHong2015" class="citation arxiv cs1">Ong, Hao Yi; Chavez, Kevin; Hong, Augustus (2015-08-18). "Distributed Deep Q-Learning". <a href="/wiki/ArXiv_(identifier)" class="mw-redirect" title="ArXiv (identifier)">arXiv</a>:<a rel="nofollow" class="external text" href="https://arxiv.org/abs/1508.04186v2">1508.04186v2</a> [<a rel="nofollow" class="external text" href="https://arxiv.org/archive/cs.LG">cs.LG</a>].</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=preprint&rft.jtitle=arXiv&rft.atitle=Distributed+Deep+Q-Learning&rft.date=2015-08-18&rft_id=info%3Aarxiv%2F1508.04186v2&rft.aulast=Ong&rft.aufirst=Hao+Yi&rft.au=Chavez%2C+Kevin&rft.au=Hong%2C+Augustus&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-DQN-158"><a href="#cite_ref-DQN_158-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFMnih2015" class="citation journal cs1">Mnih, Volodymyr; et al. (2015). "Human-level control through deep reinforcement learning". Nature. 518 (7540): 529–533. <a href="/wiki/Bibcode_(identifier)" class="mw-redirect" title="Bibcode (identifier)">Bibcode</a>:<a rel="nofollow" class="external text" href="https://ui.adsabs.harvard.edu/abs/2015Natur.518..529M">2015Natur.518..529M</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1038%2Fnature14236">10.1038/nature14236</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/25719670">25719670</a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:205242740">205242740</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Nature&rft.atitle=Human-level+control+through+deep+reinforcement+learning&rft.volume=518&rft.issue=7540&rft.pages=529-533&rft.date=2015&rft_id=info%3Adoi%2F10.1038%2Fnature14236&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A205242740%23id-name%3DS2CID&rft_id=info%3Apmid%2F25719670&rft_id=info%3Abibcode%2F2015Natur.518..529M&rft.aulast=Mnih&rft.aufirst=Volodymyr&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-159"><a href="#cite_ref-159">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFSunSessions2000" class="citation journal cs1">Sun, R.; Sessions, C. (June 2000). "Self-segmentation of sequences: automatic formation of hierarchies of sequential behaviors". IEEE Transactions on Systems, Man, and Cybernetics - Part B: Cybernetics. 30 (3): 403–418. <a href="/wiki/CiteSeerX_(identifier)" class="mw-redirect" title="CiteSeerX (identifier)">CiteSeerX</a> <a rel="nofollow" class="external text" href="https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.11.226">10.1.1.11.226</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1109%2F3477.846230">10.1109/3477.846230</a>. <a href="/wiki/ISSN_(identifier)" class="mw-redirect" title="ISSN (identifier)">ISSN</a> <a rel="nofollow" class="external text" href="https://search.worldcat.org/issn/1083-4419">1083-4419</a>. <a href="/wiki/PMID_(identifier)" class="mw-redirect" title="PMID (identifier)">PMID</a> <a rel="nofollow" class="external text" href="https://pubmed.ncbi.nlm.nih.gov/18252373">18252373</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=IEEE+Transactions+on+Systems%2C+Man%2C+and+Cybernetics+-+Part+B%3A+Cybernetics&rft.atitle=Self-segmentation+of+sequences%3A+automatic+formation+of+hierarchies+of+sequential+behaviors&rft.volume=30&rft.issue=3&rft.pages=403-418&rft.date=2000-06&rft_id=https%3A%2F%2Fciteseerx.ist.psu.edu%2Fviewdoc%2Fsummary%3Fdoi%3D10.1.1.11.226%23id-name%3DCiteSeerX&rft.issn=1083-4419&rft_id=info%3Apmid%2F18252373&rft_id=info%3Adoi%2F10.1109%2F3477.846230&rft.aulast=Sun&rft.aufirst=R.&rft.au=Sessions%2C+C.&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-CDBN-CIFAR-160"><a href="#cite_ref-CDBN-CIFAR_160-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite class="citation web cs1"><a rel="nofollow" class="external text" href="http://www.cs.toronto.edu/~kriz/conv-cifar10-aug2010.pdf">"Convolutional Deep Belief Networks on CIFAR-10"</a> (PDF). <a rel="nofollow" class="external text" href="https://web.archive.org/web/20170830060223/http://www.cs.toronto.edu/~kriz/conv-cifar10-aug2010.pdf">Archived</a> (PDF) from the original on 2017-08-30. Retrieved 2017-08-18.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=unknown&rft.btitle=Convolutional+Deep+Belief+Networks+on+CIFAR-10&rft_id=http%3A%2F%2Fwww.cs.toronto.edu%2F~kriz%2Fconv-cifar10-aug2010.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-CDBN-161"><a href="#cite_ref-CDBN_161-0">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFLeeGrosseRanganathNg2009" class="citation book cs1">Lee, Honglak; Grosse, Roger; Ranganath, Rajesh; Ng, Andrew Y. (1 January 2009). "Convolutional deep belief networks for scalable unsupervised learning of hierarchical representations". Proceedings of the 26th Annual International Conference on Machine Learning. ACM. pp. 609–616. <a href="/wiki/CiteSeerX_(identifier)" class="mw-redirect" title="CiteSeerX (identifier)">CiteSeerX</a> <a rel="nofollow" class="external text" href="https://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.149.6800">10.1.1.149.6800</a>. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1145%2F1553374.1553453">10.1145/1553374.1553453</a>. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/9781605585161" title="Special:BookSources/9781605585161"><bdi>9781605585161</bdi></a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:12008458">12008458</a>.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=bookitem&rft.atitle=Convolutional+deep+belief+networks+for+scalable+unsupervised+learning+of+hierarchical+representations&rft.btitle=Proceedings+of+the+26th+Annual+International+Conference+on+Machine+Learning&rft.pages=609-616&rft.pub=ACM&rft.date=2009-01-01&rft_id=https%3A%2F%2Fciteseerx.ist.psu.edu%2Fviewdoc%2Fsummary%3Fdoi%3D10.1.1.149.6800%23id-name%3DCiteSeerX&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A12008458%23id-name%3DS2CID&rft_id=info%3Adoi%2F10.1145%2F1553374.1553453&rft.isbn=9781605585161&rft.aulast=Lee&rft.aufirst=Honglak&rft.au=Grosse%2C+Roger&rft.au=Ranganath%2C+Rajesh&rft.au=Ng%2C+Andrew+Y.&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-162"><a href="#cite_ref-162">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFBehnke2003" class="citation book cs1">Behnke, Sven (2003). <a rel="nofollow" class="external text" href="https://www.ais.uni-bonn.de/books/LNCS2766.pdf">Hierarchical Neural Networks for Image Interpretation</a> (PDF). Lecture Notes in Computer Science. Vol. 2766. Springer. <a href="/wiki/Doi_(identifier)" class="mw-redirect" title="Doi (identifier)">doi</a>:<a rel="nofollow" class="external text" href="https://doi.org/10.1007%2Fb11963">10.1007/b11963</a>. <a href="/wiki/ISBN_(identifier)" class="mw-redirect" title="ISBN (identifier)">ISBN</a> <a href="/wiki/Special:BookSources/978-3-540-40722-5" title="Special:BookSources/978-3-540-40722-5"><bdi>978-3-540-40722-5</bdi></a>. <a href="/wiki/S2CID_(identifier)" class="mw-redirect" title="S2CID (identifier)">S2CID</a> <a rel="nofollow" class="external text" href="https://api.semanticscholar.org/CorpusID:1304548">1304548</a>. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20170810020001/http://www.ais.uni-bonn.de/books/LNCS2766.pdf">Archived</a> (PDF) from the original on 2017-08-10. Retrieved 2016-12-28.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Abook&rft.genre=book&rft.btitle=Hierarchical+Neural+Networks+for+Image+Interpretation&rft.series=Lecture+Notes+in+Computer+Science&rft.pub=Springer&rft.date=2003&rft_id=https%3A%2F%2Fapi.semanticscholar.org%2FCorpusID%3A1304548%23id-name%3DS2CID&rft_id=info%3Adoi%2F10.1007%2Fb11963&rft.isbn=978-3-540-40722-5&rft.aulast=Behnke&rft.aufirst=Sven&rft_id=https%3A%2F%2Fwww.ais.uni-bonn.de%2Fbooks%2FLNCS2766.pdf&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> <li id="cite_note-163"><a href="#cite_ref-163">^</a> <link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1238218222"><cite id="CITEREFCade_Metz2016" class="citation news cs1">Cade Metz (May 18, 2016). <a rel="nofollow" class="external text" href="https://www.wired.com/2016/05/google-tpu-custom-chips/">"Google Built Its Very Own Chips to Power Its AI Bots"</a>. Wired. <a rel="nofollow" class="external text" href="https://web.archive.org/web/20180113150305/https://www.wired.com/2016/05/google-tpu-custom-chips/">Archived</a> from the original on January 13, 2018. Retrieved March 6, 2017.</cite><span title="ctx_ver=Z39.88-2004&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.jtitle=Wired&rft.atitle=Google+Built+Its+Very+Own+Chips+to+Power+Its+AI+Bots&rft.date=2016-05-18&rft.au=Cade+Metz&rft_id=https%3A%2F%2Fwww.wired.com%2F2016%2F05%2Fgoogle-tpu-custom-chips%2F&rfr_id=info%3Asid%2Fen.wikipedia.org%3AConvolutional+neural+network" class="Z3988"> </li> </ol></div> <div class="mw-heading mw-heading2"><h2 id="External_links">External links</h2>[<a href="/w/index.php?title=Convolutional_neural_network&action=edit&section=71" title="Edit section: External links">edit</a>]</div> <ul><li><a rel="nofollow" class="external text" href="https://cs231n.github.io/">CS231n: Convolutional Neural Networks for Visual Recognition</a> — <a href="/wiki/Andrej_Karpathy" title="Andrej Karpathy">Andrej Karpathy</a>'s <a href="/wiki/Stanford_University" title="Stanford University">Stanford</a> computer science course on CNNs in computer vision</li> <li><a rel="nofollow" class="external text" href="https://github.com/vdumoulin/conv_arithmetic">vdumoulin/conv_arithmetic: A technical report on convolution arithmetic in the context of deep learning</a>. Animations of convolutions.</li></ul> <div class="navbox-styles"><link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1129693374"><style data-mw-deduplicate="TemplateStyles:r1236075235">.mw-parser-output .navbox{box-sizing:border-box;border:1px solid #a2a9b1;width:100%;clear:both;font-size:88%;text-align:center;padding:1px;margin:1em auto 0}.mw-parser-output .navbox .navbox{margin-top:0}.mw-parser-output .navbox+.navbox,.mw-parser-output .navbox+.navbox-styles+.navbox{margin-top:-1px}.mw-parser-output .navbox-inner,.mw-parser-output .navbox-subgroup{width:100%}.mw-parser-output .navbox-group,.mw-parser-output .navbox-title,.mw-parser-output .navbox-abovebelow{padding:0.25em 1em;line-height:1.5em;text-align:center}.mw-parser-output .navbox-group{white-space:nowrap;text-align:right}.mw-parser-output .navbox,.mw-parser-output .navbox-subgroup{background-color:#fdfdfd}.mw-parser-output .navbox-list{line-height:1.5em;border-color:#fdfdfd}.mw-parser-output .navbox-list-with-group{text-align:left;border-left-width:2px;border-left-style:solid}.mw-parser-output tr+tr>.navbox-abovebelow,.mw-parser-output tr+tr>.navbox-group,.mw-parser-output tr+tr>.navbox-image,.mw-parser-output tr+tr>.navbox-list{border-top:2px solid #fdfdfd}.mw-parser-output .navbox-title{background-color:#ccf}.mw-parser-output .navbox-abovebelow,.mw-parser-output .navbox-group,.mw-parser-output .navbox-subgroup .navbox-title{background-color:#ddf}.mw-parser-output .navbox-subgroup .navbox-group,.mw-parser-output .navbox-subgroup .navbox-abovebelow{background-color:#e6e6ff}.mw-parser-output .navbox-even{background-color:#f7f7f7}.mw-parser-output .navbox-odd{background-color:transparent}.mw-parser-output .navbox .hlist td dl,.mw-parser-output .navbox .hlist td ol,.mw-parser-output .navbox .hlist td ul,.mw-parser-output .navbox td.hlist dl,.mw-parser-output .navbox td.hlist ol,.mw-parser-output .navbox td.hlist ul{padding:0.125em 0}.mw-parser-output .navbox .navbar{display:block;font-size:100%}.mw-parser-output .navbox-title .navbar{float:left;text-align:left;margin-right:0.5em}body.skin--responsive .mw-parser-output .navbox-image img{max-width:none!important}@media print{body.ns-0 .mw-parser-output .navbox{display:none!important}}</style></div><div role="navigation" class="navbox" aria-labelledby="Artificial_intelligence" style="padding:3px"><table class="nowraplinks hlist mw-collapsible {{{state}}} navbox-inner" style="border-spacing:0;background:transparent;color:inherit"><tbody><tr><th scope="col" class="navbox-title" colspan="2"><link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1129693374"><link rel="mw-deduplicated-inline-style" href="mw-data:TemplateStyles:r1239400231"><div class="navbar plainlinks hlist navbar-mini"><ul><li class="nv-view"><a href="/wiki/Template:Artificial_intelligence_navbox" title="Template:Artificial intelligence navbox"><abbr title="View this template">v</abbr></a></li><li class="nv-talk"><a href="/wiki/Template_talk:Artificial_intelligence_navbox" title="Template talk:Artificial intelligence navbox"><abbr title="Discuss this template">t</abbr></a></li><li class="nv-edit"><a href="/wiki/Special:EditPage/Template:Artificial_intelligence_navbox" title="Special:EditPage/Template:Artificial intelligence navbox"><abbr title="Edit this template">e</abbr></a></li></ul></div><div id="Artificial_intelligence" style="font-size:114%;margin:0 4em"><a href="/wiki/Artificial_intelligence" title="Artificial intelligence">Artificial intelligence</a></div></th></tr><tr><th scope="row" class="navbox-group" style="width:1%">Concepts</th><td class="navbox-list-with-group navbox-list navbox-odd" style="width:100%;padding:0"><div style="padding:0 0.25em"> <ul><li><a href="/wiki/Parameter" title="Parameter">Parameter</a> <ul><li><a href="/wiki/Hyperparameter_(machine_learning)" title="Hyperparameter (machine learning)">Hyperparameter</a></li></ul></li> <li><a href="/wiki/Loss_functions_for_classification" title="Loss functions for classification">Loss functions</a></li> <li><a href="/wiki/Regression_analysis" title="Regression analysis">Regression</a> <ul><li><a href="/wiki/Bias%E2%80%93variance_tradeoff" title="Bias–variance tradeoff">Bias–variance tradeoff</a></li> <li><a href="/wiki/Double_descent" title="Double descent">Double descent</a></li> <li><a href="/wiki/Overfitting" title="Overfitting">Overfitting</a></li></ul></li> <li><a href="/wiki/Cluster_analysis" title="Cluster analysis">Clustering</a></li> <li><a href="/wiki/Gradient_descent" title="Gradient descent">Gradient descent</a> <ul><li><a href="/wiki/Stochastic_gradient_descent" title="Stochastic gradient descent">SGD</a></li> <li><a href="/wiki/Quasi-Newton_method" title="Quasi-Newton method">Quasi-Newton method</a></li> <li><a href="/wiki/Conjugate_gradient_method" title="Conjugate gradient method">Conjugate gradient method</a></li></ul></li> <li><a href="/wiki/Backpropagation" title="Backpropagation">Backpropagation</a></li> <li><a href="/wiki/Attention_(machine_learning)" title="Attention (machine learning)">Attention</a></li> <li><a href="/wiki/Convolution" title="Convolution">Convolution</a></li> <li><a href="/wiki/Normalization_(machine_learning)" title="Normalization (machine learning)">Normalization</a> <ul><li><a href="/wiki/Batch_normalization" title="Batch normalization">Batchnorm</a></li></ul></li> <li><a href="/wiki/Activation_function" title="Activation function">Activation</a> <ul><li><a href="/wiki/Softmax_function" title="Softmax function">Softmax</a></li> <li><a href="/wiki/Sigmoid_function" title="Sigmoid function">Sigmoid</a></li> <li><a href="/wiki/Rectifier_(neural_networks)" title="Rectifier (neural networks)">Rectifier</a></li></ul></li> <li><a href="/wiki/Gating_mechanism" title="Gating mechanism">Gating</a></li> <li><a href="/wiki/Weight_initialization" title="Weight initialization">Weight initialization</a></li> <li><a href="/wiki/Regularization_(mathematics)" title="Regularization (mathematics)">Regularization</a></li> <li><a href="/wiki/Training,_validation,_and_test_data_sets" title="Training, validation, and test data sets">Datasets</a> <ul><li><a href="/wiki/Data_augmentation" title="Data augmentation">Augmentation</a></li></ul></li> <li><a href="/wiki/Reinforcement_learning" title="Reinforcement learning">Reinforcement learning</a> <ul><li><a href="/wiki/Q-learning" title="Q-learning">Q-learning</a></li> <li><a href="/wiki/State%E2%80%93action%E2%80%93reward%E2%80%93state%E2%80%93action" title="State–action–reward–state–action">SARSA</a></li> <li><a href="/wiki/Imitation_learning" title="Imitation learning">Imitation</a></li></ul></li> <li><a href="/wiki/Diffusion_process" title="Diffusion process">Diffusion</a></li> <li><a href="/wiki/Autoregressive_model" title="Autoregressive model">Autoregression</a></li> <li><a href="/wiki/Adversarial_machine_learning" title="Adversarial machine learning">Adversary</a></li> <li><a href="/wiki/Hallucination_(artificial_intelligence)" title="Hallucination (artificial intelligence)">Hallucination</a></li></ul> </div></td></tr><tr><th scope="row" class="navbox-group" style="width:1%">Applications</th><td class="navbox-list-with-group navbox-list navbox-even" style="width:100%;padding:0"><div style="padding:0 0.25em"> <ul><li><a href="/wiki/Machine_learning" title="Machine learning">Machine learning</a> <ul><li><a href="/wiki/Prompt_engineering#In-context_learning" title="Prompt engineering">In-context learning</a></li></ul></li> <li><a href="/wiki/Neural_network_(machine_learning)" title="Neural network (machine learning)">Artificial neural network</a> <ul><li><a href="/wiki/Deep_learning" title="Deep learning">Deep learning</a></li></ul></li> <li><a href="/wiki/Language_model" title="Language model">Language model</a> <ul><li><a href="/wiki/Large_language_model" title="Large language model">Large language model</a></li> <li><a href="/wiki/Neural_machine_translation" title="Neural machine translation">NMT</a></li></ul></li> <li><a href="/wiki/Artificial_general_intelligence" title="Artificial general intelligence">Artificial general intelligence</a></li></ul> </div></td></tr><tr><th scope="row" class="navbox-group" style="width:1%">Implementations</th><td class="navbox-list-with-group navbox-list navbox-odd" style="width:100%;padding:0"><div style="padding:0 0.25em"></div><table class="nowraplinks navbox-subgroup" style="border-spacing:0"><tbody><tr><th scope="row" class="navbox-group" style="width:1%">Audio–visual</th><td class="navbox-list-with-group navbox-list navbox-odd" style="width:100%;padding:0"><div style="padding:0 0.25em"> <ul><li><a href="/wiki/AlexNet" title="AlexNet">AlexNet</a></li> <li><a href="/wiki/WaveNet" title="WaveNet">WaveNet</a></li> <li><a href="/wiki/Human_image_synthesis" title="Human image synthesis">Human image synthesis</a></li> <li><a href="/wiki/Handwriting_recognition" title="Handwriting recognition">HWR</a></li> <li><a href="/wiki/Optical_character_recognition" title="Optical character recognition">OCR</a></li> <li><a href="/wiki/Deep_learning_speech_synthesis" title="Deep learning speech synthesis">Speech synthesis</a> <ul><li><a href="/wiki/ElevenLabs" title="ElevenLabs">ElevenLabs</a></li></ul></li> <li><a href="/wiki/Speech_recognition" title="Speech recognition">Speech recognition</a></li> <li><a href="/wiki/Facial_recognition_system" title="Facial recognition system">Facial recognition</a></li> <li><a href="/wiki/AlphaFold" title="AlphaFold">AlphaFold</a></li> <li><a href="/wiki/Text-to-image_model" title="Text-to-image model">Text-to-image models</a> <ul><li><a href="/wiki/Latent_diffusion_model" title="Latent diffusion model">Latent diffusion model</a></li> <li><a href="/wiki/DALL-E" title="DALL-E">DALL-E</a></li> <li><a href="/wiki/Flux_(text-to-image_model)" title="Flux (text-to-image model)">Flux</a></li> <li><a href="/wiki/Ideogram_(text-to-image_model)" title="Ideogram (text-to-image model)">Ideogram</a></li> <li><a href="/wiki/Midjourney" title="Midjourney">Midjourney</a></li> <li><a href="/wiki/Stable_Diffusion" title="Stable Diffusion">Stable Diffusion</a></li></ul></li> <li><a href="/wiki/Text-to-video_model" title="Text-to-video model">Text-to-video models</a> <ul><li><a href="/wiki/Sora_(text-to-video_model)" title="Sora (text-to-video model)">Sora</a></li> <li><a href="/wiki/Dream_Machine_(text-to-video_model)" title="Dream Machine (text-to-video model)">Dream Machine</a></li> <li><a href="/wiki/VideoPoet" title="VideoPoet">VideoPoet</a></li></ul></li> <li><a href="/wiki/Whisper_(speech_recognition_system)" title="Whisper (speech recognition system)">Whisper</a></li></ul> </div></td></tr><tr><th scope="row" class="navbox-group" style="width:1%">Text</th><td class="navbox-list-with-group navbox-list navbox-even" style="width:100%;padding:0"><div style="padding:0 0.25em"> <ul><li><a href="/wiki/Word2vec" title="Word2vec">Word2vec</a></li> <li><a href="/wiki/Seq2seq" title="Seq2seq">Seq2seq</a></li> <li><a href="/wiki/GloVe" title="GloVe">GloVe</a></li> <li><a href="/wiki/BERT_(language_model)" title="BERT (language model)">BERT</a></li> <li><a href="/wiki/T5_(language_model)" title="T5 (language model)">T5</a></li> <li><a href="/wiki/Llama_(language_model)" title="Llama (language model)">Llama</a></li> <li><a href="/wiki/Chinchilla_(language_model)" title="Chinchilla (language model)">Chinchilla AI</a></li> <li><a href="/wiki/PaLM" title="PaLM">PaLM</a></li> <li><a href="/wiki/Generative_pre-trained_transformer" title="Generative pre-trained transformer">GPT</a> <ul><li><a href="/wiki/GPT-1" title="GPT-1">1</a></li> <li><a href="/wiki/GPT-J" title="GPT-J">J</a></li> <li><a href="/wiki/GPT-2" title="GPT-2">2</a></li> <li><a href="/wiki/GPT-3" title="GPT-3">3</a></li> <li><a href="/wiki/ChatGPT" title="ChatGPT">ChatGPT</a></li> <li><a href="/wiki/GPT-4" title="GPT-4">4</a></li> <li><a href="/wiki/GPT-4o" title="GPT-4o">4o</a></li> <li><a href="/wiki/OpenAI_o1" title="OpenAI o1">o1</a></li></ul></li> <li><a href="/wiki/Claude_(language_model)" title="Claude (language model)">Claude</a></li> <li><a href="/wiki/Gemini_(language_model)" title="Gemini (language model)">Gemini</a></li> <li><a href="/wiki/Grok_(chatbot)" title="Grok (chatbot)">Grok</a></li> <li><a href="/wiki/LaMDA" title="LaMDA">LaMDA</a></li> <li><a href="/wiki/BLOOM_(language_model)" title="BLOOM (language model)">BLOOM</a></li> <li><a href="/wiki/Project_Debater" title="Project Debater">Project Debater</a></li> <li><a href="/wiki/IBM_Watson" title="IBM Watson">IBM Watson</a></li> <li><a href="/wiki/IBM_Watsonx" title="IBM Watsonx">IBM Watsonx</a></li> <li><a href="/wiki/IBM_Granite" title="IBM Granite">Granite</a></li> <li><a href="/wiki/Huawei_PanGu" title="Huawei PanGu">PanGu-Σ</a></li></ul> </div></td></tr><tr><th scope="row" class="navbox-group" style="width:1%">Decisional</th><td class="navbox-list-with-group navbox-list navbox-odd" style="width:100%;padding:0"><div style="padding:0 0.25em"> <ul><li><a href="/wiki/AlphaGo" title="AlphaGo">AlphaGo</a></li> <li><a href="/wiki/AlphaZero" title="AlphaZero">AlphaZero</a></li> <li><a href="/wiki/OpenAI_Five" title="OpenAI Five">OpenAI Five</a></li> <li><a href="/wiki/Self-driving_car" title="Self-driving car">Self-driving car</a></li> <li><a href="/wiki/MuZero" title="MuZero">MuZero</a></li> <li><a href="/wiki/Action_selection" title="Action selection">Action selection</a> <ul><li><a href="/wiki/AutoGPT" title="AutoGPT">AutoGPT</a></li></ul></li> <li><a href="/wiki/Robot_control" title="Robot control">Robot control</a></li></ul> </div></td></tr></tbody></table><div></div></td></tr><tr><th scope="row" class="navbox-group" style="width:1%">People</th><td class="navbox-list-with-group navbox-list navbox-even" style="width:100%;padding:0"><div style="padding:0 0.25em"> <ul><li><a href="/wiki/Alan_Turing" title="Alan Turing">Alan Turing</a></li> <li><a href="/wiki/Claude_Shannon" title="Claude Shannon">Claude Shannon</a></li> <li><a href="/wiki/Allen_Newell" title="Allen Newell">Allen Newell</a></li> <li><a href="/wiki/Herbert_A._Simon" title="Herbert A. Simon">Herbert A. Simon</a></li> <li><a href="/wiki/Frank_Rosenblatt" title="Frank Rosenblatt">Frank Rosenblatt</a></li> <li><a href="/wiki/Marvin_Minsky" title="Marvin Minsky">Marvin Minsky</a></li> <li><a href="/wiki/John_McCarthy_(computer_scientist)" title="John McCarthy (computer scientist)">John McCarthy</a></li> <li><a href="/wiki/Nathaniel_Rochester_(computer_scientist)" title="Nathaniel Rochester (computer scientist)">Nathaniel Rochester</a></li> <li><a href="/wiki/Seymour_Papert" title="Seymour Papert">Seymour Papert</a></li> <li><a href="/wiki/Joseph_Weizenbaum" title="Joseph Weizenbaum">Joseph Weizenbaum</a></li> <li><a href="/wiki/Bernard_Widrow" title="Bernard Widrow">Bernard Widrow</a></li> <li><a href="/wiki/Paul_Werbos" title="Paul Werbos">Paul Werbos</a></li> <li><a href="/wiki/Yoshua_Bengio" title="Yoshua Bengio">Yoshua Bengio</a></li> <li><a href="/wiki/Alex_Graves_(computer_scientist)" title="Alex Graves (computer scientist)">Alex Graves</a></li> <li><a href="/wiki/Ian_Goodfellow" title="Ian Goodfellow">Ian Goodfellow</a></li> <li><a href="/wiki/Stephen_Grossberg" title="Stephen Grossberg">Stephen Grossberg</a></li> <li><a href="/wiki/Demis_Hassabis" title="Demis Hassabis">Demis Hassabis</a></li> <li><a href="/wiki/Geoffrey_Hinton" title="Geoffrey Hinton">Geoffrey Hinton</a></li> <li><a href="/wiki/Yann_LeCun" title="Yann LeCun">Yann LeCun</a></li> <li><a href="/wiki/Fei-Fei_Li" title="Fei-Fei Li">Fei-Fei Li</a></li> <li><a href="/wiki/Andrew_Ng" title="Andrew Ng">Andrew Ng</a></li> <li><a href="/wiki/J%C3%BCrgen_Schmidhuber" title="Jürgen Schmidhuber">Jürgen Schmidhuber</a></li> <li><a href="/wiki/David_Silver_(computer_scientist)" title="David Silver (computer scientist)">David Silver</a></li> <li><a href="/wiki/Ilya_Sutskever" title="Ilya Sutskever">Ilya Sutskever</a></li></ul> </div></td></tr><tr><th scope="row" class="navbox-group" style="width:1%">Organizations</th><td class="navbox-list-with-group navbox-list navbox-odd" style="width:100%;padding:0"><div style="padding:0 0.25em"> <ul><li><a href="/wiki/Anthropic" title="Anthropic">Anthropic</a></li> <li><a href="/wiki/EleutherAI" title="EleutherAI">EleutherAI</a></li> <li><a href="/wiki/Google_DeepMind" title="Google DeepMind">Google DeepMind</a></li> <li><a href="/wiki/Hugging_Face" title="Hugging Face">Hugging Face</a></li> <li><a href="/wiki/Kuaishou" title="Kuaishou">Kuaishou</a></li> <li><a href="/wiki/Meta_AI" title="Meta AI">Meta AI</a></li> <li><a href="/wiki/Mila_(research_institute)" title="Mila (research institute)">Mila</a></li> <li><a href="/wiki/MiniMax_(company)" title="MiniMax (company)">MiniMax</a></li> <li><a href="/wiki/Mistral_AI" title="Mistral AI">Mistral AI</a></li> <li><a href="/wiki/MIT_Computer_Science_and_Artificial_Intelligence_Laboratory" title="MIT Computer Science and Artificial Intelligence Laboratory">MIT CSAIL</a></li> <li><a href="/wiki/OpenAI" title="OpenAI">OpenAI</a></li> <li><a href="/wiki/Runway_(company)" title="Runway (company)">Runway</a></li> <li><a href="/wiki/XAI_(company)" title="XAI (company)">xAI</a></li></ul> </div></td></tr><tr><th scope="row" class="navbox-group" style="width:1%">Architectures</th><td class="navbox-list-with-group navbox-list navbox-even" style="width:100%;padding:0"><div style="padding:0 0.25em"> <ul><li><a href="/wiki/Neural_Turing_machine" title="Neural Turing machine">Neural Turing machine</a></li> <li><a href="/wiki/Differentiable_neural_computer" title="Differentiable neural computer">Differentiable neural computer</a></li> <li><a href="/wiki/Transformer_(deep_learning_architecture)" title="Transformer (deep learning architecture)">Transformer</a> <ul><li><a href="/wiki/Vision_transformer" title="Vision transformer">Vision transformer (ViT)</a></li></ul></li> <li><a href="/wiki/Recurrent_neural_network" title="Recurrent neural network">Recurrent neural network (RNN)</a></li> <li><a href="/wiki/Long_short-term_memory" title="Long short-term memory">Long short-term memory (LSTM)</a></li> <li><a href="/wiki/Gated_recurrent_unit" title="Gated recurrent unit">Gated recurrent unit (GRU)</a></li> <li><a href="/wiki/Echo_state_network" title="Echo state network">Echo state network</a></li> <li><a href="/wiki/Multilayer_perceptron" title="Multilayer perceptron">Multilayer perceptron (MLP)</a></li> <li><a class="mw-selflink selflink">Convolutional neural network (CNN)</a></li> <li><a href="/wiki/Residual_neural_network" title="Residual neural network">Residual neural network (RNN)</a></li> <li><a href="/wiki/Highway_network" title="Highway network">Highway network</a></li> <li><a href="/wiki/Mamba_(deep_learning_architecture)" title="Mamba (deep learning architecture)">Mamba</a></li> <li><a href="/wiki/Autoencoder" title="Autoencoder">Autoencoder</a></li> <li><a href="/wiki/Variational_autoencoder" title="Variational autoencoder">Variational autoencoder (VAE)</a></li> <li><a href="/wiki/Generative_adversarial_network" title="Generative adversarial network">Generative adversarial network (GAN)</a></li> <li><a href="/wiki/Graph_neural_network" title="Graph neural network">Graph neural network (GNN)</a></li></ul> </div></td></tr><tr><td class="navbox-abovebelow" colspan="2"><div> <ul><li><a href="/wiki/File:Symbol_portal_class.svg" class="mw-file-description" title="Portal"><img alt="" src="//upload.wikimedia.org/wikipedia/en/thumb/e/e2/Symbol_portal_class.svg/16px-Symbol_portal_class.svg.png" decoding="async" width="16" height="16" class="mw-file-element" srcset="//upload.wikimedia.org/wikipedia/en/thumb/e/e2/Symbol_portal_class.svg/23px-Symbol_portal_class.svg.png 1.5x, //upload.wikimedia.org/wikipedia/en/thumb/e/e2/Symbol_portal_class.svg/31px-Symbol_portal_class.svg.png 2x" data-file-width="180" data-file-height="185" /></a> Portals <ul><li><a href="/wiki/Portal:Technology" title="Portal:Technology">Technology</a></li></ul></li> <li><img alt="" src="//upload.wikimedia.org/wikipedia/en/thumb/9/96/Symbol_category_class.svg/16px-Symbol_category_class.svg.png" decoding="async" width="16" height="16" class="mw-file-element" srcset="//upload.wikimedia.org/wikipedia/en/thumb/9/96/Symbol_category_class.svg/23px-Symbol_category_class.svg.png 1.5x, //upload.wikimedia.org/wikipedia/en/thumb/9/96/Symbol_category_class.svg/31px-Symbol_category_class.svg.png 2x" data-file-width="180" data-file-height="185" /> Categories <ul><li><a href="/wiki/Category:Artificial_neural_networks" title="Category:Artificial neural networks">Artificial neural networks</a></li> <li><a href="/wiki/Category:Machine_learning" title="Category:Machine learning">Machine learning</a></li></ul></li></ul> </div></td></tr></tbody></table></div>    </div><noscript><img src="https://login.wikimedia.org/wiki/Special:CentralAutoLogin/start?type=1x1" alt="" width="1" height="1" style="border: none; position: absolute;"></noscript> <div class="printfooter" data-nosnippet="">Retrieved from "<a dir="ltr" href="https://en.wikipedia.org/w/index.php?title=Convolutional_neural_network&oldid=1257441256">https://en.wikipedia.org/w/index.php?title=Convolutional_neural_network&oldid=1257441256</a>"</div></div> <div id="catlinks" class="catlinks" data-mw="interface"><div id="mw-normal-catlinks" class="mw-normal-catlinks"><a href="/wiki/Help:Category" title="Help:Category">Categories</a>: <ul><li><a href="/wiki/Category:Neural_network_architectures" title="Category:Neural network architectures">Neural network architectures</a></li><li><a href="/wiki/Category:Computer_vision" title="Category:Computer vision">Computer vision</a></li><li><a href="/wiki/Category:Computational_neuroscience" title="Category:Computational neuroscience">Computational neuroscience</a></li></ul></div><div id="mw-hidden-catlinks" class="mw-hidden-catlinks mw-hidden-cats-hidden">Hidden categories: <ul><li><a href="/wiki/Category:CS1_maint:_location_missing_publisher" title="Category:CS1 maint: location missing publisher">CS1 maint: location missing publisher</a></li><li><a href="/wiki/Category:CS1_maint:_multiple_names:_authors_list" title="Category:CS1 maint: multiple names: authors list">CS1 maint: multiple names: authors list</a></li><li><a href="/wiki/Category:CS1:_long_volume_value" title="Category:CS1: long volume value">CS1: long volume value</a></li><li><a href="/wiki/Category:CS1_Japanese-language_sources_(ja)" title="Category:CS1 Japanese-language sources (ja)">CS1 Japanese-language sources (ja)</a></li><li><a href="/wiki/Category:Webarchive_template_wayback_links" title="Category:Webarchive template wayback links">Webarchive template wayback links</a></li><li><a href="/wiki/Category:All_articles_with_dead_external_links" title="Category:All articles with dead external links">All articles with dead external links</a></li><li><a href="/wiki/Category:Articles_with_dead_external_links_from_July_2022" title="Category:Articles with dead external links from July 2022">Articles with dead external links from July 2022</a></li><li><a href="/wiki/Category:CS1_German-language_sources_(de)" title="Category:CS1 German-language sources (de)">CS1 German-language sources (de)</a></li><li><a href="/wiki/Category:Articles_with_short_description" title="Category:Articles with short description">Articles with short description</a></li><li><a href="/wiki/Category:Short_description_is_different_from_Wikidata" title="Category:Short description is different from Wikidata">Short description is different from Wikidata</a></li><li><a href="/wiki/Category:Articles_needing_additional_references_from_June_2019" title="Category:Articles needing additional references from June 2019">Articles needing additional references from June 2019</a></li><li><a href="/wiki/Category:All_articles_needing_additional_references" title="Category:All articles needing additional references">All articles needing additional references</a></li><li><a href="/wiki/Category:All_articles_with_specifically_marked_weasel-worded_phrases" title="Category:All articles with specifically marked weasel-worded phrases">All articles with specifically marked weasel-worded phrases</a></li><li><a href="/wiki/Category:Articles_with_specifically_marked_weasel-worded_phrases_from_April_2023" title="Category:Articles with specifically marked weasel-worded phrases from April 2023">Articles with specifically marked weasel-worded phrases from April 2023</a></li><li><a href="/wiki/Category:All_articles_with_unsourced_statements" title="Category:All articles with unsourced statements">All articles with unsourced statements</a></li><li><a href="/wiki/Category:Articles_with_unsourced_statements_from_October_2017" title="Category:Articles with unsourced statements from October 2017">Articles with unsourced statements from October 2017</a></li><li><a href="/wiki/Category:Articles_containing_explicitly_cited_British_English-language_text" title="Category:Articles containing explicitly cited British English-language text">Articles containing explicitly cited British English-language text</a></li><li><a href="/wiki/Category:All_articles_needing_examples" title="Category:All articles needing examples">All articles needing examples</a></li><li><a href="/wiki/Category:Articles_needing_examples_from_October_2017" title="Category:Articles needing examples from October 2017">Articles needing examples from October 2017</a></li><li><a href="/wiki/Category:Articles_with_unsourced_statements_from_March_2024" title="Category:Articles with unsourced statements from March 2024">Articles with unsourced statements from March 2024</a></li><li><a href="/wiki/Category:Articles_with_specifically_marked_weasel-worded_phrases_from_December_2018" title="Category:Articles with specifically marked weasel-worded phrases from December 2018">Articles with specifically marked weasel-worded phrases from December 2018</a></li><li><a href="/wiki/Category:Articles_needing_additional_references_from_June_2017" title="Category:Articles needing additional references from June 2017">Articles needing additional references from June 2017</a></li><li><a href="/wiki/Category:Wikipedia_articles_needing_clarification_from_December_2018" title="Category:Wikipedia articles needing clarification from December 2018">Wikipedia articles needing clarification from December 2018</a></li><li><a href="/wiki/Category:Articles_with_unsourced_statements_from_June_2019" title="Category:Articles with unsourced statements from June 2019">Articles with unsourced statements from June 2019</a></li></ul></div></div> </div> </main> </div> <div class="mw-footer-container"> <footer id="footer" class="mw-footer" > <ul id="footer-info"> <li id="footer-info-lastmod"> This page was last edited on 14 November 2024, at 22:28 (UTC).</li> <li id="footer-info-copyright">Text is available under the <a href="/wiki/Wikipedia:Text_of_the_Creative_Commons_Attribution-ShareAlike_4.0_International_License" title="Wikipedia:Text of the Creative Commons Attribution-ShareAlike 4.0 International License">Creative Commons Attribution-ShareAlike 4.0 License</a>; additional terms may apply. By using this site, you agree to the <a href="https://foundation.wikimedia.org/wiki/Special:MyLanguage/Policy:Terms_of_Use" class="extiw" title="foundation:Special:MyLanguage/Policy:Terms of Use">Terms of Use</a> and <a href="https://foundation.wikimedia.org/wiki/Special:MyLanguage/Policy:Privacy_policy" class="extiw" title="foundation:Special:MyLanguage/Policy:Privacy policy">Privacy Policy</a>. Wikipedia® is a registered trademark of the <a rel="nofollow" class="external text" href="https://wikimediafoundation.org/">Wikimedia Foundation, Inc.</a>, a non-profit organization.</li> </ul> <ul id="footer-places"> <li id="footer-places-privacy"><a href="https://foundation.wikimedia.org/wiki/Special:MyLanguage/Policy:Privacy_policy">Privacy policy</a></li> <li id="footer-places-about"><a href="/wiki/Wikipedia:About">About Wikipedia</a></li> <li id="footer-places-disclaimers"><a href="/wiki/Wikipedia:General_disclaimer">Disclaimers</a></li> <li id="footer-places-contact"><a href="//en.wikipedia.org/wiki/Wikipedia:Contact_us">Contact Wikipedia</a></li> <li id="footer-places-wm-codeofconduct"><a href="https://foundation.wikimedia.org/wiki/Special:MyLanguage/Policy:Universal_Code_of_Conduct">Code of Conduct</a></li> <li id="footer-places-developers"><a href="https://developer.wikimedia.org">Developers</a></li> <li id="footer-places-statslink"><a href="https://stats.wikimedia.org/#/en.wikipedia.org">Statistics</a></li> <li id="footer-places-cookiestatement"><a href="https://foundation.wikimedia.org/wiki/Special:MyLanguage/Policy:Cookie_statement">Cookie statement</a></li> <li id="footer-places-mobileview"><a href="//en.m.wikipedia.org/w/index.php?title=Convolutional_neural_network&mobileaction=toggle_view_mobile" class="noprint stopMobileRedirectToggle">Mobile view</a></li> </ul> <ul id="footer-icons" class="noprint"> <li id="footer-copyrightico"><a href="https://wikimediafoundation.org/" class="cdx-button cdx-button--fake-button cdx-button--size-large cdx-button--fake-button--enabled"><img src="/static/images/footer/wikimedia-button.svg" width="84" height="29" alt="Wikimedia Foundation" loading="lazy"></a></li> <li id="footer-poweredbyico"><a href="https://www.mediawiki.org/" class="cdx-button cdx-button--fake-button cdx-button--size-large cdx-button--fake-button--enabled"><img src="/w/resources/assets/poweredby_mediawiki.svg" alt="Powered by MediaWiki" width="88" height="31" loading="lazy"></a></li> </ul> </footer> </div> </div> </div> <div class="vector-settings" id="p-dock-bottom"> <ul></ul> </div><script>(RLQ=window.RLQ||[]).push(function(){mw.config.set({"wgHostname":"mw-web.codfw.main-f69cdc8f6-vbjf2","wgBackendResponseTime":201,"wgPageParseReport":{"limitreport":{"cputime":"2.352","walltime":"2.704","ppvisitednodes":{"value":13267,"limit":1000000},"postexpandincludesize":{"value":506786,"limit":2097152},"templateargumentsize":{"value":12000,"limit":2097152},"expansiondepth":{"value":18,"limit":100},"expensivefunctioncount":{"value":21,"limit":500},"unstrip-depth":{"value":1,"limit":20},"unstrip-size":{"value":625082,"limit":5000000},"entityaccesscount":{"value":1,"limit":400},"timingprofile":["100.00% 2282.813 1 -total"," 60.46% 1380.166 2 Template:Reflist"," 25.62% 584.792 58 Template:Cite_journal"," 9.02% 205.817 26 Template:Cite_book"," 8.12% 185.341 29 Template:Cite_arXiv"," 6.90% 157.559 1 Template:Short_description"," 5.55% 126.604 2 Template:Pagetype"," 5.34% 121.936 2 Template:Langx"," 5.34% 121.853 1 Template:Machine_learning"," 5.06% 115.574 1 Template:Sidebar_with_collapsible_lists"]},"scribunto":{"limitreport-timeusage":{"value":"1.524","limit":"10.000"},"limitreport-memusage":{"value":21119599,"limit":52428800},"limitreport-profile":[["MediaWiki\\Extension\\Scribunto\\Engines\\LuaSandbox\\LuaSandboxCallback::callParserFunction","300","19.2"],["?","200","12.8"],["MediaWiki\\Extension\\Scribunto\\Engines\\LuaSandbox\\LuaSandboxCallback::gsub","180","11.5"],["\u003Cmw.lua:694\u003E","100","6.4"],["MediaWiki\\Extension\\Scribunto\\Engines\\LuaSandbox\\LuaSandboxCallback::match","100","6.4"],["dataWrapper \u003Cmw.lua:672\u003E","100","6.4"],["recursiveClone \u003CmwInit.lua:45\u003E","80","5.1"],["MediaWiki\\Extension\\Scribunto\\Engines\\LuaSandbox\\LuaSandboxCallback::sub","60","3.8"],["key_to_lower \u003CModule:Lang/data:22\u003E","40","2.6"],["MediaWiki\\Extension\\Scribunto\\Engines\\LuaSandbox\\LuaSandboxCallback::find","40","2.6"],["[others]","360","23.1"]]},"cachereport":{"origin":"mw-web.codfw.main-f69cdc8f6-g5pcr","timestamp":"20241122145448","ttl":2592000,"transientcontent":false}}});});</script> <script type="application/ld+json">{"@context":"https:\/\/schema.org","@type":"Article","name":"Convolutional neural network","url":"https:\/\/en.wikipedia.org\/wiki\/Convolutional_neural_network","sameAs":"http:\/\/www.wikidata.org\/entity\/Q17084460","mainEntity":"http:\/\/www.wikidata.org\/entity\/Q17084460","author":{"@type":"Organization","name":"Contributors to Wikimedia projects"},"publisher":{"@type":"Organization","name":"Wikimedia Foundation, Inc.","logo":{"@type":"ImageObject","url":"https:\/\/www.wikimedia.org\/static\/images\/wmf-hor-googpub.png"}},"datePublished":"2013-08-31T15:20:31Z","dateModified":"2024-11-14T22:28:17Z","headline":"regularized type of feed-forward neural network that learns features by itself via filter (or kernel) optimization"}</script> </body> </html>