Search | arXiv e-print repository

<!DOCTYPE html> <html lang="en"> <head> <meta charset="utf-8"/> <meta name="viewport" content="width=device-width, initial-scale=1"/>  <link rel="apple-touch-icon" sizes="180x180" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/apple-touch-icon.png"> <link rel="icon" type="image/png" sizes="32x32" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/favicon-32x32.png"> <link rel="icon" type="image/png" sizes="16x16" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/favicon-16x16.png"> <link rel="manifest" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/site.webmanifest"> <link rel="mask-icon" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/safari-pinned-tab.svg" color="#b31b1b"> <link rel="shortcut icon" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/favicon.ico"> <meta name="msapplication-TileColor" content="#b31b1b"> <meta name="msapplication-config" content="images/icons/browserconfig.xml"> <meta name="theme-color" content="#b31b1b">  <title>Search | arXiv e-print repository</title> <script defer src="https://static.arxiv.org/static/base/1.0.0a5/fontawesome-free-5.11.2-web/js/all.js"></script> <link rel="stylesheet" href="https://static.arxiv.org/static/base/1.0.0a5/css/arxivstyle.css" /> <script type="text/x-mathjax-config"> MathJax.Hub.Config({ messageStyle: "none", extensions: ["tex2jax.js"], jax: ["input/TeX", "output/HTML-CSS"], tex2jax: { inlineMath: [ ['$','$'], ["\$","\$"] ], displayMath: [ ['$$','$$'], ["\\[","\\]"] ], processEscapes: true, ignoreClass: '.*', processClass: 'mathjax.*' }, TeX: { extensions: ["AMSmath.js", "AMSsymbols.js", "noErrors.js"], noErrors: { inlineDelimiters: ["$","$"], multiLine: false, style: { "font-size": "normal", "border": "" } } }, "HTML-CSS": { availableFonts: ["TeX"] } }); </script> <script src='//static.arxiv.org/MathJax-2.7.3/MathJax.js'></script> <script src="https://static.arxiv.org/static/base/1.0.0a5/js/notification.js"></script> <link rel="stylesheet" href="https://static.arxiv.org/static/search/0.5.6/css/bulma-tooltip.min.css" /> <link rel="stylesheet" href="https://static.arxiv.org/static/search/0.5.6/css/search.css" /> <script src="https://code.jquery.com/jquery-3.2.1.slim.min.js" integrity="sha256-k2WSCIexGzOj3Euiig+TlR8gA0EmPjuc79OEeY5L45g=" crossorigin="anonymous"></script> <script src="https://static.arxiv.org/static/search/0.5.6/js/fieldset.js"></script> <style> radio#cf-customfield_11400 { display: none; } </style> </head> <body> <header><a href="#main-container" class="is-sr-only">Skip to main content</a>  <div class="attribution level is-marginless" role="banner"> <div class="level-left"> <a class="level-item" href="https://cornell.edu/"><img src="https://static.arxiv.org/static/base/1.0.0a5/images/cornell-reduced-white-SMALL.svg" alt="Cornell University" width="200" aria-label="logo" /></a> </div> <div class="level-right is-marginless"><p class="sponsors level-item is-marginless"><span id="support-ack-url">We gratefully acknowledge support from<br /> the Simons Foundation, <a href="https://info.arxiv.org/about/ourmembers.html">member institutions</a>, and all contributors. <a href="https://info.arxiv.org/about/donate.html">Donate</a></span></p></div> </div>  <div class="identity level is-marginless"> <div class="level-left"> <div class="level-item"> <a class="arxiv" href="https://arxiv.org/" aria-label="arxiv-logo"> <img src="https://static.arxiv.org/static/base/1.0.0a5/images/arxiv-logo-one-color-white.svg" aria-label="logo" alt="arxiv logo" width="85" style="width:85px;"/> </a> </div> </div> <div class="search-block level-right"> <form class="level-item mini-search" method="GET" action="https://arxiv.org/search"> <div class="field has-addons"> <div class="control"> <input class="input is-small" type="text" name="query" placeholder="Search..." aria-label="Search term or terms" /> <p class="help"><a href="https://info.arxiv.org/help">Help</a> | <a href="https://arxiv.org/search/advanced">Advanced Search</a></p> </div> <div class="control"> <div class="select is-small"> <select name="searchtype" aria-label="Field to search"> <option value="all" selected="selected">All fields</option> <option value="title">Title</option> <option value="author">Author</option> <option value="abstract">Abstract</option> <option value="comments">Comments</option> <option value="journal_ref">Journal reference</option> <option value="acm_class">ACM classification</option> <option value="msc_class">MSC classification</option> <option value="report_num">Report number</option> <option value="paper_id">arXiv identifier</option> <option value="doi">DOI</option> <option value="orcid">ORCID</option> <option value="author_id">arXiv author ID</option> <option value="help">Help pages</option> <option value="full_text">Full text</option> </select> </div> </div> <input type="hidden" name="source" value="header"> <button class="button is-small is-cul-darker">Search</button> </div> </form> </div> </div>  <div class="container"> <div class="user-tools is-size-7 has-text-right has-text-weight-bold" role="navigation" aria-label="User menu"> <a href="https://arxiv.org/login">Login</a> </div> </div> </header> <main class="container" id="main-container"> <div class="level is-marginless"> <div class="level-left"> <h1 class="title is-clearfix"> Showing 1–50 of 174 results for author: <span class="mathjax">Shin, Y</span> </h1> </div> <div class="level-right is-hidden-mobile">  <span class="help" style="display: inline-block;"><a href="https://github.com/arXiv/arxiv-search/releases">Search v0.5.6 released 2020-02-24</a>  </span> </div> </div> <div class="content"> <form method="GET" action="/search/cs" aria-role="search"> Searching in archive <strong>cs</strong>. <a href="/search/?searchtype=author&query=Shin%2C+Y">Search in all archives.</a> <div class="field has-addons-tablet"> <div class="control is-expanded"> <label for="query" class="hidden-label">Search term or terms</label> <input class="input is-medium" id="query" name="query" placeholder="Search term..." type="text" value="Shin, Y"> </div> <div class="select control is-medium"> <label class="is-hidden" for="searchtype">Field</label> <select class="is-medium" id="searchtype" name="searchtype"><option value="all">All fields</option><option value="title">Title</option><option selected value="author">Author(s)</option><option value="abstract">Abstract</option><option value="comments">Comments</option><option value="journal_ref">Journal reference</option><option value="acm_class">ACM classification</option><option value="msc_class">MSC classification</option><option value="report_num">Report number</option><option value="paper_id">arXiv identifier</option><option value="doi">DOI</option><option value="orcid">ORCID</option><option value="license">License (URI)</option><option value="author_id">arXiv author ID</option><option value="help">Help pages</option><option value="full_text">Full text</option></select> </div> <div class="control"> <button class="button is-link is-medium">Search</button> </div> </div> <div class="field"> <div class="control is-size-7"> <label class="radio"> <input checked id="abstracts-0" name="abstracts" type="radio" value="show"> Show abstracts </label> <label class="radio"> <input id="abstracts-1" name="abstracts" type="radio" value="hide"> Hide abstracts </label> </div> </div> <div class="is-clearfix" style="height: 2.5em"> <div class="is-pulled-right"> <a href="/search/advanced?terms-0-term=Shin%2C+Y&terms-0-field=author&size=50&order=-announced_date_first">Advanced Search</a> </div> </div> <input type="hidden" name="order" value="-announced_date_first"> <input type="hidden" name="size" value="50"> </form> <div class="level breathe-horizontal"> <div class="level-left"> <form method="GET" action="/search/"> <div style="display: none;"> <select id="searchtype" name="searchtype"><option value="all">All fields</option><option value="title">Title</option><option selected value="author">Author(s)</option><option value="abstract">Abstract</option><option value="comments">Comments</option><option value="journal_ref">Journal reference</option><option value="acm_class">ACM classification</option><option value="msc_class">MSC classification</option><option value="report_num">Report number</option><option value="paper_id">arXiv identifier</option><option value="doi">DOI</option><option value="orcid">ORCID</option><option value="license">License (URI)</option><option value="author_id">arXiv author ID</option><option value="help">Help pages</option><option value="full_text">Full text</option></select> <input id="query" name="query" type="text" value="Shin, Y"> <ul id="abstracts"><li><input checked id="abstracts-0" name="abstracts" type="radio" value="show"> <label for="abstracts-0">Show abstracts</label></li><li><input id="abstracts-1" name="abstracts" type="radio" value="hide"> <label for="abstracts-1">Hide abstracts</label></li></ul> </div> <div class="box field is-grouped is-grouped-multiline level-item"> <div class="control"> <span class="select is-small"> <select id="size" name="size"><option value="25">25</option><option selected value="50">50</option><option value="100">100</option><option value="200">200</option></select> </span> <label for="size">results per page</label>. </div> <div class="control"> <label for="order">Sort results by</label> <span class="select is-small"> <select id="order" name="order"><option selected value="-announced_date_first">Announcement date (newest first)</option><option value="announced_date_first">Announcement date (oldest first)</option><option value="-submitted_date">Submission date (newest first)</option><option value="submitted_date">Submission date (oldest first)</option><option value="">Relevance</option></select> </span> </div> <div class="control"> <button class="button is-small is-link">Go</button> </div> </div> </form> </div> </div> <nav class="pagination is-small is-centered breathe-horizontal" role="navigation" aria-label="pagination"> <a href="" class="pagination-previous is-invisible">Previous </a> <a href="/search/?searchtype=author&query=Shin%2C+Y&start=50" class="pagination-next" >Next </a> <ul class="pagination-list"> <li> <a href="/search/?searchtype=author&query=Shin%2C+Y&start=0" class="pagination-link is-current" aria-label="Goto page 1">1 </a> </li> <li> <a href="/search/?searchtype=author&query=Shin%2C+Y&start=50" class="pagination-link " aria-label="Page 2" aria-current="page">2 </a> </li> <li> <a href="/search/?searchtype=author&query=Shin%2C+Y&start=100" class="pagination-link " aria-label="Page 3" aria-current="page">3 </a> </li> <li> <a href="/search/?searchtype=author&query=Shin%2C+Y&start=150" class="pagination-link " aria-label="Page 4" aria-current="page">4 </a> </li> </ul> </nav> <ol class="breathe-horizontal" start="1"> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.12220">arXiv:2411.12220</a> <span> [<a href="https://arxiv.org/pdf/2411.12220">pdf</a>, <a href="https://arxiv.org/format/2411.12220">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Cryptography and Security">cs.CR</span> </div> </div> <p class="title is-5 mathjax"> DeTrigger: A Gradient-Centric Approach to Backdoor Attack Mitigation in Federated Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lee%2C+K">Kichang Lee</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yujin Shin</a>, <a href="/search/cs?searchtype=author&query=Yun%2C+J">Jonghyuk Yun</a>, <a href="/search/cs?searchtype=author&query=Han%2C+J">Jun Han</a>, <a href="/search/cs?searchtype=author&query=Ko%2C+J">JeongGil Ko</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.12220v1-abstract-short" style="display: inline;"> Federated Learning (FL) enables collaborative model training across distributed devices while preserving local data privacy, making it ideal for mobile and embedded systems. However, the decentralized nature of FL also opens vulnerabilities to model poisoning attacks, particularly backdoor attacks, where adversaries implant trigger patterns to manipulate model predictions. In this paper, we propos… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.12220v1-abstract-full').style.display = 'inline'; document.getElementById('2411.12220v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.12220v1-abstract-full" style="display: none;"> Federated Learning (FL) enables collaborative model training across distributed devices while preserving local data privacy, making it ideal for mobile and embedded systems. However, the decentralized nature of FL also opens vulnerabilities to model poisoning attacks, particularly backdoor attacks, where adversaries implant trigger patterns to manipulate model predictions. In this paper, we propose DeTrigger, a scalable and efficient backdoor-robust federated learning framework that leverages insights from adversarial attack methodologies. By employing gradient analysis with temperature scaling, DeTrigger detects and isolates backdoor triggers, allowing for precise model weight pruning of backdoor activations without sacrificing benign model knowledge. Extensive evaluations across four widely used datasets demonstrate that DeTrigger achieves up to 251x faster detection than traditional methods and mitigates backdoor attacks by up to 98.9%, with minimal impact on global model accuracy. Our findings establish DeTrigger as a robust and scalable solution to protect federated learning environments against sophisticated backdoor threats. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.12220v1-abstract-full').style.display = 'none'; document.getElementById('2411.12220v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 18 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">14 pages</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">MSC Class:</span> 68T07 <span class="has-text-black-bis has-text-weight-semibold">ACM Class:</span> I.2.11 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.08626">arXiv:2411.08626</a> <span> [<a href="https://arxiv.org/pdf/2411.08626">pdf</a>, <a href="https://arxiv.org/format/2411.08626">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Software Engineering">cs.SE</span> </div> </div> <p class="title is-5 mathjax"> Learning-Guided Fuzzing for Testing Stateful SDN Controllers </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Ollando%2C+R">Rapha毛l Ollando</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+S+Y">Seung Yeob Shin</a>, <a href="/search/cs?searchtype=author&query=Briand%2C+L+C">Lionel C. Briand</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2411.08626v1-abstract-short" style="display: inline;"> Controllers for software-defined networks (SDNs) are centralised software components that enable advanced network functionalities, such as dynamic traffic engineering and network virtualisation. However, these functionalities increase the complexity of SDN controllers, making thorough testing crucial. SDN controllers are stateful, interacting with multiple network devices through sequences of cont… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.08626v1-abstract-full').style.display = 'inline'; document.getElementById('2411.08626v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.08626v1-abstract-full" style="display: none;"> Controllers for software-defined networks (SDNs) are centralised software components that enable advanced network functionalities, such as dynamic traffic engineering and network virtualisation. However, these functionalities increase the complexity of SDN controllers, making thorough testing crucial. SDN controllers are stateful, interacting with multiple network devices through sequences of control messages. Identifying stateful failures in an SDN controller is challenging due to the infinite possible sequences of control messages, which result in an unbounded number of stateful interactions between the controller and network devices. In this article, we propose SeqFuzzSDN, a learning-guided fuzzing method for testing stateful SDN controllers. SeqFuzzSDN aims to (1) efficiently explore the state space of the SDN controller under test, (2) generate effective and diverse tests (i.e., control message sequences) to uncover failures, and (3) infer accurate failure-inducing models that characterise the message sequences leading to failures. In addition, we compare SeqFuzzSDN with three extensions of state-of-the-art (SOTA) methods for fuzzing SDNs. Our findings show that, compared to the extended SOTA methods, SeqFuzzSDN (1) generates more diverse message sequences that lead to failures within the same time budget, and (2) produces more accurate failure-inducing models, significantly outperforming the other extended SOTA methods in terms of sensitivity. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.08626v1-abstract-full').style.display = 'none'; document.getElementById('2411.08626v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 13 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2410.12827">arXiv:2410.12827</a> <span> [<a href="https://arxiv.org/pdf/2410.12827">pdf</a>, <a href="https://arxiv.org/format/2410.12827">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Image and Video Processing">eess.IV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> DyMix: Dynamic Frequency Mixup Scheduler based Unsupervised Domain Adaptation for Enhancing Alzheimer's Disease Identification </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yooseung Shin</a>, <a href="/search/cs?searchtype=author&query=Oh%2C+K">Kwanseok Oh</a>, <a href="/search/cs?searchtype=author&query=Suk%2C+H">Heung-Il Suk</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2410.12827v1-abstract-short" style="display: inline;"> Advances in deep learning (DL)-based models for brain image analysis have significantly enhanced the accuracy of Alzheimer's disease (AD) diagnosis, allowing for more timely interventions. Despite these advancements, most current DL models suffer from performance degradation when inferring on unseen domain data owing to the variations in data distributions, a phenomenon known as domain shift. To a… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.12827v1-abstract-full').style.display = 'inline'; document.getElementById('2410.12827v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2410.12827v1-abstract-full" style="display: none;"> Advances in deep learning (DL)-based models for brain image analysis have significantly enhanced the accuracy of Alzheimer's disease (AD) diagnosis, allowing for more timely interventions. Despite these advancements, most current DL models suffer from performance degradation when inferring on unseen domain data owing to the variations in data distributions, a phenomenon known as domain shift. To address this challenge, we propose a novel approach called the dynamic frequency mixup scheduler (DyMix) for unsupervised domain adaptation. Contrary to the conventional mixup technique, which involves simple linear interpolations between predefined data points from the frequency space, our proposed DyMix dynamically adjusts the magnitude of the frequency regions being mixed from the source and target domains. Such an adaptive strategy optimizes the model's capacity to deal with domain variability, thereby enhancing its generalizability across the target domain. In addition, we incorporate additional strategies to further enforce the model's robustness against domain shifts, including leveraging amplitude-phase recombination to ensure resilience to intensity variations and applying self-adversarial learning to derive domain-invariant feature representations. Experimental results on two benchmark datasets quantitatively and qualitatively validated the effectiveness of our DyMix in that we demonstrated its outstanding performance in AD diagnosis compared to state-of-the-art methods. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.12827v1-abstract-full').style.display = 'none'; document.getElementById('2410.12827v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 2 October, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">10 pages, 5 figures, 3 tables</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2410.12561">arXiv:2410.12561</a> <span> [<a href="https://arxiv.org/pdf/2410.12561">pdf</a>, <a href="https://arxiv.org/format/2410.12561">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Development of Image Collection Method Using YOLO and Siamese Network </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Shin%2C+C+Y">Chan Young Shin</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+A+H">Ah Hyun Lee</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+J+Y">Jun Young Lee</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+J+M">Ji Min Lee</a>, <a href="/search/cs?searchtype=author&query=Park%2C+S+J">Soo Jin Park</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2410.12561v1-abstract-short" style="display: inline;"> As we enter the era of big data, collecting high-quality data is very important. However, collecting data by humans is not only very time-consuming but also expensive. Therefore, many scientists have devised various methods to collect data using computers. Among them, there is a method called web crawling, but the authors found that the crawling method has a problem in that unintended data is coll… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.12561v1-abstract-full').style.display = 'inline'; document.getElementById('2410.12561v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2410.12561v1-abstract-full" style="display: none;"> As we enter the era of big data, collecting high-quality data is very important. However, collecting data by humans is not only very time-consuming but also expensive. Therefore, many scientists have devised various methods to collect data using computers. Among them, there is a method called web crawling, but the authors found that the crawling method has a problem in that unintended data is collected along with the user. The authors found that this can be filtered using the object recognition model YOLOv10. However, there are cases where data that is not properly filtered remains. Here, image reclassification was performed by additionally utilizing the distance output from the Siamese network, and higher performance was recorded than other classification models. (average \_f1 score YOLO+MobileNet 0.678->YOLO+SiameseNet 0.772)) The user can specify a distance threshold to adjust the balance between data deficiency and noise-robustness. The authors also found that the Siamese network can achieve higher performance with fewer resources because the cropped images are used for object recognition when processing images in the Siamese network. (Class 20 mean-based f1 score, non-crop+Siamese(MobileNetV3-Small) 80.94 -> crop preprocessing+Siamese(MobileNetV3-Small) 82.31) In this way, the image retrieval system that utilizes two consecutive models to reduce errors can save users' time and effort, and build better quality data faster and with fewer resources than before. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.12561v1-abstract-full').style.display = 'none'; document.getElementById('2410.12561v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 16 October, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">15 pages, 13 figures, 2 tables</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2409.19840">arXiv:2409.19840</a> <span> [<a href="https://arxiv.org/pdf/2409.19840">pdf</a>, <a href="https://arxiv.org/format/2409.19840">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> Textual Training for the Hassle-Free Removal of Unwanted Visual Data: Case Studies on OOD and Hateful Image Detection </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lee%2C+S">Saehyung Lee</a>, <a href="/search/cs?searchtype=author&query=Mok%2C+J">Jisoo Mok</a>, <a href="/search/cs?searchtype=author&query=Park%2C+S">Sangha Park</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yongho Shin</a>, <a href="/search/cs?searchtype=author&query=Jung%2C+D">Dahuin Jung</a>, <a href="/search/cs?searchtype=author&query=Yoon%2C+S">Sungroh Yoon</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2409.19840v2-abstract-short" style="display: inline;"> In our study, we explore methods for detecting unwanted content lurking in visual datasets. We provide a theoretical analysis demonstrating that a model capable of successfully partitioning visual data can be obtained using only textual data. Based on the analysis, we propose Hassle-Free Textual Training (HFTT), a streamlined method capable of acquiring detectors for unwanted visual content, using… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2409.19840v2-abstract-full').style.display = 'inline'; document.getElementById('2409.19840v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2409.19840v2-abstract-full" style="display: none;"> In our study, we explore methods for detecting unwanted content lurking in visual datasets. We provide a theoretical analysis demonstrating that a model capable of successfully partitioning visual data can be obtained using only textual data. Based on the analysis, we propose Hassle-Free Textual Training (HFTT), a streamlined method capable of acquiring detectors for unwanted visual content, using only synthetic textual data in conjunction with pre-trained vision-language models. HFTT features an innovative objective function that significantly reduces the necessity for human involvement in data annotation. Furthermore, HFTT employs a clever textual data synthesis method, effectively emulating the integration of unknown visual data distribution into the training process at no extra cost. The unique characteristics of HFTT extend its utility beyond traditional out-of-distribution detection, making it applicable to tasks that address more abstract concepts. We complement our analyses with experiments in out-of-distribution detection and hateful image detection. Our codes are available at https://github.com/Saehyung-Lee/HFTT <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2409.19840v2-abstract-full').style.display = 'none'; document.getElementById('2409.19840v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 23 October, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 29 September, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">NeurIPS 2024</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2407.13942">arXiv:2407.13942</a> <span> [<a href="https://arxiv.org/pdf/2407.13942">pdf</a>, <a href="https://arxiv.org/format/2407.13942">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computers and Society">cs.CY</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computation and Language">cs.CL</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Social and Information Networks">cs.SI</span> </div> </div> <p class="title is-5 mathjax"> Harmful Suicide Content Detection </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Park%2C+K">Kyumin Park</a>, <a href="/search/cs?searchtype=author&query=Baik%2C+M+J">Myung Jae Baik</a>, <a href="/search/cs?searchtype=author&query=Hwang%2C+Y">YeongJun Hwang</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yen Shin</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+H">HoJae Lee</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+R">Ruda Lee</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+S+M">Sang Min Lee</a>, <a href="/search/cs?searchtype=author&query=Sun%2C+J+Y+H">Je Young Hannah Sun</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+A+R">Ah Rah Lee</a>, <a href="/search/cs?searchtype=author&query=Yoon%2C+S+Y">Si Yeun Yoon</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+D">Dong-ho Lee</a>, <a href="/search/cs?searchtype=author&query=Moon%2C+J">Jihyung Moon</a>, <a href="/search/cs?searchtype=author&query=Bak%2C+J">JinYeong Bak</a>, <a href="/search/cs?searchtype=author&query=Cho%2C+K">Kyunghyun Cho</a>, <a href="/search/cs?searchtype=author&query=Paik%2C+J">Jong-Woo Paik</a>, <a href="/search/cs?searchtype=author&query=Park%2C+S">Sungjoon Park</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2407.13942v1-abstract-short" style="display: inline;"> Harmful suicide content on the Internet is a significant risk factor inducing suicidal thoughts and behaviors among vulnerable populations. Despite global efforts, existing resources are insufficient, specifically in high-risk regions like the Republic of Korea. Current research mainly focuses on understanding negative effects of such content or suicide risk in individuals, rather than on automati… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2407.13942v1-abstract-full').style.display = 'inline'; document.getElementById('2407.13942v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2407.13942v1-abstract-full" style="display: none;"> Harmful suicide content on the Internet is a significant risk factor inducing suicidal thoughts and behaviors among vulnerable populations. Despite global efforts, existing resources are insufficient, specifically in high-risk regions like the Republic of Korea. Current research mainly focuses on understanding negative effects of such content or suicide risk in individuals, rather than on automatically detecting the harmfulness of content. To fill this gap, we introduce a harmful suicide content detection task for classifying online suicide content into five harmfulness levels. We develop a multi-modal benchmark and a task description document in collaboration with medical professionals, and leverage large language models (LLMs) to explore efficient methods for moderating such content. Our contributions include proposing a novel detection task, a multi-modal Korean benchmark with expert annotations, and suggesting strategies using LLMs to detect illegal and harmful content. Owing to the potential harm involved, we publicize our implementations and benchmark, incorporating an ethical verification process. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2407.13942v1-abstract-full').style.display = 'none'; document.getElementById('2407.13942v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 2 June, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">30 pages, 7 figures</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2407.06441">arXiv:2407.06441</a> <span> [<a href="https://arxiv.org/pdf/2407.06441">pdf</a>, <a href="https://arxiv.org/format/2407.06441">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Human-Computer Interaction">cs.HC</span> </div> </div> <p class="title is-5 mathjax"> A Study of Digital Appliances Accessibility for People with Visual Disabilities </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=An%2C+H">Hyunjin An</a>, <a href="/search/cs?searchtype=author&query=Kim%2C+H">Hyundoug Kim</a>, <a href="/search/cs?searchtype=author&query=Hong%2C+S">Seungwoo Hong</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Youngsun Shin</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2407.06441v1-abstract-short" style="display: inline;"> This research aims to find where visually impaired users find appliances hard to use and suggest guideline to solve this issue. 181 visually impaired users have been surveyed, and 12 visually impaired users have been selected based on disability cause and classification. In a home-like environment, we had participants perform tasks which were sorted using Hierarchical task analysis on six major ho… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2407.06441v1-abstract-full').style.display = 'inline'; document.getElementById('2407.06441v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2407.06441v1-abstract-full" style="display: none;"> This research aims to find where visually impaired users find appliances hard to use and suggest guideline to solve this issue. 181 visually impaired users have been surveyed, and 12 visually impaired users have been selected based on disability cause and classification. In a home-like environment, we had participants perform tasks which were sorted using Hierarchical task analysis on six major home appliances. From this research we found out that home appliances sometimes only provide visual information which causes difficulty in sensory processing. Also, interfaces tactile/auditory feedbacks are the same making it hard for people to recognize which feature is processed. Blind users cannot see the provided information so they rely on long-term memory to use products. This research provides guideline for button, knob and remote control interface for visually impaired users. This information will be helpful for project planners, designers, and developers to create products which are accessible by visually impaired people. Some of the features will be applied to upcoming home appliance products. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2407.06441v1-abstract-full').style.display = 'none'; document.getElementById('2407.06441v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 8 July, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">10 pages, 3 figures</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">MSC Class:</span> 68U35 <span class="has-text-black-bis has-text-weight-semibold">ACM Class:</span> D.2.2 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2407.05527">arXiv:2407.05527</a> <span> [<a href="https://arxiv.org/pdf/2407.05527">pdf</a>, <a href="https://arxiv.org/format/2407.05527">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Image and Video Processing">eess.IV</span> </div> </div> <p class="title is-5 mathjax"> Rethinking Image Skip Connections in StyleGAN2 </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Park%2C+S">Seung Park</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yong-Goo Shin</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2407.05527v1-abstract-short" style="display: inline;"> Various models based on StyleGAN have gained significant traction in the field of image synthesis, attributed to their robust training stability and superior performances. Within the StyleGAN framework, the adoption of image skip connection is favored over the traditional residual connection. However, this preference is just based on empirical observations; there has not been any in-depth mathemat… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2407.05527v1-abstract-full').style.display = 'inline'; document.getElementById('2407.05527v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2407.05527v1-abstract-full" style="display: none;"> Various models based on StyleGAN have gained significant traction in the field of image synthesis, attributed to their robust training stability and superior performances. Within the StyleGAN framework, the adoption of image skip connection is favored over the traditional residual connection. However, this preference is just based on empirical observations; there has not been any in-depth mathematical analysis on it yet. To rectify this situation, this brief aims to elucidate the mathematical meaning of the image skip connection and introduce a groundbreaking methodology, termed the image squeeze connection, which significantly improves the quality of image synthesis. Specifically, we analyze the image skip connection technique to reveal its problem and introduce the proposed method which not only effectively boosts the GAN performance but also reduces the required number of network parameters. Extensive experiments on various datasets demonstrate that the proposed method consistently enhances the performance of state-of-the-art models based on StyleGAN. We believe that our findings represent a vital advancement in the field of image synthesis, suggesting a novel direction for future research and applications. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2407.05527v1-abstract-full').style.display = 'none'; document.getElementById('2407.05527v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 7 July, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2407.03086">arXiv:2407.03086</a> <span> [<a href="https://arxiv.org/pdf/2407.03086">pdf</a>, <a href="https://arxiv.org/format/2407.03086">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Distributed, Parallel, and Cluster Computing">cs.DC</span> </div> </div> <p class="title is-5 mathjax"> Effective Heterogeneous Federated Learning via Efficient Hypernetwork-based Weight Generation </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yujin Shin</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+K">Kichang Lee</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+S">Sungmin Lee</a>, <a href="/search/cs?searchtype=author&query=Choi%2C+Y+R">You Rim Choi</a>, <a href="/search/cs?searchtype=author&query=Kim%2C+H">Hyung-Sin Kim</a>, <a href="/search/cs?searchtype=author&query=Ko%2C+J">JeongGil Ko</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2407.03086v2-abstract-short" style="display: inline;"> While federated learning leverages distributed client resources, it faces challenges due to heterogeneous client capabilities. This necessitates allocating models suited to clients' resources and careful parameter aggregation to accommodate this heterogeneity. We propose HypeMeFed, a novel federated learning framework for supporting client heterogeneity by combining a multi-exit network architectu… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2407.03086v2-abstract-full').style.display = 'inline'; document.getElementById('2407.03086v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2407.03086v2-abstract-full" style="display: none;"> While federated learning leverages distributed client resources, it faces challenges due to heterogeneous client capabilities. This necessitates allocating models suited to clients' resources and careful parameter aggregation to accommodate this heterogeneity. We propose HypeMeFed, a novel federated learning framework for supporting client heterogeneity by combining a multi-exit network architecture with hypernetwork-based model weight generation. This approach aligns the feature spaces of heterogeneous model layers and resolves per-layer information disparity during weight aggregation. To practically realize HypeMeFed, we also propose a low-rank factorization approach to minimize computation and memory overhead associated with hypernetworks. Our evaluations on a real-world heterogeneous device testbed indicate that \system enhances accuracy by 5.12% over FedAvg, reduces the hypernetwork memory requirements by 98.22%, and accelerates its operations by 1.86x compared to a naive hypernetwork approach. These results demonstrate HypeMeFed's effectiveness in leveraging and engaging heterogeneous clients for federated learning. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2407.03086v2-abstract-full').style.display = 'none'; document.getElementById('2407.03086v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 3 October, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 3 July, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2406.14308">arXiv:2406.14308</a> <span> [<a href="https://arxiv.org/pdf/2406.14308">pdf</a>, <a href="https://arxiv.org/format/2406.14308">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Image and Video Processing">eess.IV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> </div> </div> <p class="title is-5 mathjax"> FIESTA: Fourier-Based Semantic Augmentation with Uncertainty Guidance for Enhanced Domain Generalizability in Medical Image Segmentation </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Oh%2C+K">Kwanseok Oh</a>, <a href="/search/cs?searchtype=author&query=Jeon%2C+E">Eunjin Jeon</a>, <a href="/search/cs?searchtype=author&query=Heo%2C+D">Da-Woon Heo</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yooseung Shin</a>, <a href="/search/cs?searchtype=author&query=Suk%2C+H">Heung-Il Suk</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2406.14308v1-abstract-short" style="display: inline;"> Single-source domain generalization (SDG) in medical image segmentation (MIS) aims to generalize a model using data from only one source domain to segment data from an unseen target domain. Despite substantial advances in SDG with data augmentation, existing methods often fail to fully consider the details and uncertain areas prevalent in MIS, leading to mis-segmentation. This paper proposes a Fou… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2406.14308v1-abstract-full').style.display = 'inline'; document.getElementById('2406.14308v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2406.14308v1-abstract-full" style="display: none;"> Single-source domain generalization (SDG) in medical image segmentation (MIS) aims to generalize a model using data from only one source domain to segment data from an unseen target domain. Despite substantial advances in SDG with data augmentation, existing methods often fail to fully consider the details and uncertain areas prevalent in MIS, leading to mis-segmentation. This paper proposes a Fourier-based semantic augmentation method called FIESTA using uncertainty guidance to enhance the fundamental goals of MIS in an SDG context by manipulating the amplitude and phase components in the frequency domain. The proposed Fourier augmentative transformer addresses semantic amplitude modulation based on meaningful angular points to induce pertinent variations and harnesses the phase spectrum to ensure structural coherence. Moreover, FIESTA employs epistemic uncertainty to fine-tune the augmentation process, improving the ability of the model to adapt to diverse augmented data and concentrate on areas with higher ambiguity. Extensive experiments across three cross-domain scenarios demonstrate that FIESTA surpasses recent state-of-the-art SDG approaches in segmentation performance and significantly contributes to boosting the applicability of the model in medical imaging modalities. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2406.14308v1-abstract-full').style.display = 'none'; document.getElementById('2406.14308v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 20 June, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">40 pages, 7 figures, 5 tables</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2406.11504">arXiv:2406.11504</a> <span> [<a href="https://arxiv.org/pdf/2406.11504">pdf</a>, <a href="https://arxiv.org/format/2406.11504">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Information Theory">cs.IT</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Neural and Evolutionary Computing">cs.NE</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Social and Information Networks">cs.SI</span> </div> </div> <p class="title is-5 mathjax"> On the Feasibility of Fidelity$^-$ for Graph Pruning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yong-Min Shin</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+W">Won-Yong Shin</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2406.11504v1-abstract-short" style="display: inline;"> As one of popular quantitative metrics to assess the quality of explanation of graph neural networks (GNNs), fidelity measures the output difference after removing unimportant parts of the input graph. Fidelity has been widely used due to its straightforward interpretation that the underlying model should produce similar predictions when features deemed unimportant from the explanation are removed… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2406.11504v1-abstract-full').style.display = 'inline'; document.getElementById('2406.11504v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2406.11504v1-abstract-full" style="display: none;"> As one of popular quantitative metrics to assess the quality of explanation of graph neural networks (GNNs), fidelity measures the output difference after removing unimportant parts of the input graph. Fidelity has been widely used due to its straightforward interpretation that the underlying model should produce similar predictions when features deemed unimportant from the explanation are removed. This raises a natural question: "Does fidelity induce a global (soft) mask for graph pruning?" To solve this, we aim to explore the potential of the fidelity measure to be used for graph pruning, eventually enhancing the GNN models for better efficiency. To this end, we propose Fidelity$^-$-inspired Pruning (FiP), an effective framework to construct global edge masks from local explanations. Our empirical observations using 7 edge attribution methods demonstrate that, surprisingly, general eXplainable AI methods outperform methods tailored to GNNs in terms of graph pruning performance. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2406.11504v1-abstract-full').style.display = 'none'; document.getElementById('2406.11504v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 17 June, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">6 pages, 3 figures, 2 tables; IJCAI Workshop on Explainable AI (XAI 2024) (to appear) (Please cite our workshop version.)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2406.08051">arXiv:2406.08051</a> <span> [<a href="https://arxiv.org/pdf/2406.08051">pdf</a>, <a href="https://arxiv.org/format/2406.08051">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Hardware Architecture">cs.AR</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Performance">cs.PF</span> </div> </div> <p class="title is-5 mathjax"> ONNXim: A Fast, Cycle-level Multi-core NPU Simulator </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Ham%2C+H">Hyungkyu Ham</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+W">Wonhyuk Yang</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yunseon Shin</a>, <a href="/search/cs?searchtype=author&query=Woo%2C+O">Okkyun Woo</a>, <a href="/search/cs?searchtype=author&query=Heo%2C+G">Guseul Heo</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+S">Sangyeop Lee</a>, <a href="/search/cs?searchtype=author&query=Park%2C+J">Jongse Park</a>, <a href="/search/cs?searchtype=author&query=Kim%2C+G">Gwangsun Kim</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2406.08051v1-abstract-short" style="display: inline;"> As DNNs are widely adopted in various application domains while demanding increasingly higher compute and memory requirements, designing efficient and performant NPUs (Neural Processing Units) is becoming more important. However, existing architectural NPU simulators lack support for high-speed simulation, multi-core modeling, multi-tenant scenarios, detailed DRAM/NoC modeling, and/or different de… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2406.08051v1-abstract-full').style.display = 'inline'; document.getElementById('2406.08051v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2406.08051v1-abstract-full" style="display: none;"> As DNNs are widely adopted in various application domains while demanding increasingly higher compute and memory requirements, designing efficient and performant NPUs (Neural Processing Units) is becoming more important. However, existing architectural NPU simulators lack support for high-speed simulation, multi-core modeling, multi-tenant scenarios, detailed DRAM/NoC modeling, and/or different deep learning frameworks. To address these limitations, this work proposes ONNXim, a fast cycle-level simulator for multi-core NPUs in DNN serving systems. It takes DNN models represented in the ONNX graph format generated from various deep learning frameworks for ease of simulation. In addition, based on the observation that typical NPU cores process tensor tiles from on-chip scratchpad memory with deterministic compute latency, we forgo a detailed modeling for the computation while still preserving simulation accuracy. ONNXim also preserves dependencies between compute and tile DMAs. Meanwhile, the DRAM and NoC are modeled in cycle-level to properly model contention among multiple cores that can execute different DNN models for multi-tenancy. Consequently, ONNXim is significantly faster than existing simulators (e.g., by up to 384x over Accel-sim) and enables various case studies, such as multi-tenant NPUs, that were previously impractical due to slow speed and/or lack of functionalities. ONNXim is publicly available at https://github.com/PSAL-POSTECH/ONNXim. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2406.08051v1-abstract-full').style.display = 'none'; document.getElementById('2406.08051v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 12 June, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2406.04612">arXiv:2406.04612</a> <span> [<a href="https://arxiv.org/pdf/2406.04612">pdf</a>, <a href="https://arxiv.org/format/2406.04612">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Information Theory">cs.IT</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Neural and Evolutionary Computing">cs.NE</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Social and Information Networks">cs.SI</span> </div> </div> <p class="title is-5 mathjax"> Revisiting Attention Weights as Interpretations of Message-Passing Neural Networks </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yong-Min Shin</a>, <a href="/search/cs?searchtype=author&query=Li%2C+S">Siqing Li</a>, <a href="/search/cs?searchtype=author&query=Cao%2C+X">Xin Cao</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+W">Won-Yong Shin</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2406.04612v1-abstract-short" style="display: inline;"> The self-attention mechanism has been adopted in several widely-used message-passing neural networks (MPNNs) (e.g., GATs), which adaptively controls the amount of information that flows along the edges of the underlying graph. This usage of attention has made such models a baseline for studies on explainable AI (XAI) since interpretations via attention have been popularized in various domains (e.g… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2406.04612v1-abstract-full').style.display = 'inline'; document.getElementById('2406.04612v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2406.04612v1-abstract-full" style="display: none;"> The self-attention mechanism has been adopted in several widely-used message-passing neural networks (MPNNs) (e.g., GATs), which adaptively controls the amount of information that flows along the edges of the underlying graph. This usage of attention has made such models a baseline for studies on explainable AI (XAI) since interpretations via attention have been popularized in various domains (e.g., natural language processing and computer vision). However, existing studies often use naive calculations to derive attribution scores from attention, and do not take the precise and careful calculation of edge attribution into consideration. In our study, we aim to fill the gap between the widespread usage of attention-enabled MPNNs and their potential in largely under-explored explainability, a topic that has been actively investigated in other areas. To this end, as the first attempt, we formalize the problem of edge attribution from attention weights in GNNs. Then, we propose GATT, an edge attribution calculation method built upon the computation tree. Through comprehensive experiments, we demonstrate the effectiveness of our proposed method when evaluating attributions from GATs. Conversely, we empirically validate that simply averaging attention weights over graph attention layers is insufficient to interpret the GAT model's behavior. Code is publicly available at https://github.com/jordan7186/GAtt/tree/main. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2406.04612v1-abstract-full').style.display = 'none'; document.getElementById('2406.04612v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 6 June, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">11 pages, 3 figures, 5 tables</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2405.02845">arXiv:2405.02845</a> <span> [<a href="https://arxiv.org/pdf/2405.02845">pdf</a>, <a href="https://arxiv.org/format/2405.02845">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Molecular Networks">q-bio.MN</span> </div> </div> <p class="title is-5 mathjax"> Data-Efficient Molecular Generation with Hierarchical Textual Inversion </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Kim%2C+S">Seojin Kim</a>, <a href="/search/cs?searchtype=author&query=Nam%2C+J">Jaehyun Nam</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+S">Sihyun Yu</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Younghoon Shin</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+J">Jinwoo Shin</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2405.02845v3-abstract-short" style="display: inline;"> Developing an effective molecular generation framework even with a limited number of molecules is often important for its practical deployment, e.g., drug discovery, since acquiring task-related molecular data requires expensive and time-consuming experimental costs. To tackle this issue, we introduce Hierarchical textual Inversion for Molecular generation (HI-Mol), a novel data-efficient molecula… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2405.02845v3-abstract-full').style.display = 'inline'; document.getElementById('2405.02845v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2405.02845v3-abstract-full" style="display: none;"> Developing an effective molecular generation framework even with a limited number of molecules is often important for its practical deployment, e.g., drug discovery, since acquiring task-related molecular data requires expensive and time-consuming experimental costs. To tackle this issue, we introduce Hierarchical textual Inversion for Molecular generation (HI-Mol), a novel data-efficient molecular generation method. HI-Mol is inspired by the importance of hierarchical information, e.g., both coarse- and fine-grained features, in understanding the molecule distribution. We propose to use multi-level embeddings to reflect such hierarchical features based on the adoption of the recent textual inversion technique in the visual domain, which achieves data-efficient image generation. Compared to the conventional textual inversion method in the image domain using a single-level token embedding, our multi-level token embeddings allow the model to effectively learn the underlying low-shot molecule distribution. We then generate molecules based on the interpolation of the multi-level token embeddings. Extensive experiments demonstrate the superiority of HI-Mol with notable data-efficiency. For instance, on QM9, HI-Mol outperforms the prior state-of-the-art method with 50x less training data. We also show the effectiveness of molecules generated by HI-Mol in low-shot molecular property prediction. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2405.02845v3-abstract-full').style.display = 'none'; document.getElementById('2405.02845v3-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 16 July, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 5 May, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">ICML 2024</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2404.19381">arXiv:2404.19381</a> <span> [<a href="https://arxiv.org/pdf/2404.19381">pdf</a>, <a href="https://arxiv.org/format/2404.19381">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Hardware Architecture">cs.AR</span> </div> </div> <p class="title is-5 mathjax"> Low-overhead General-purpose Near-Data Processing in CXL Memory Expanders </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Ham%2C+H">Hyungkyu Ham</a>, <a href="/search/cs?searchtype=author&query=Hong%2C+J">Jeongmin Hong</a>, <a href="/search/cs?searchtype=author&query=Park%2C+G">Geonwoo Park</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yunseon Shin</a>, <a href="/search/cs?searchtype=author&query=Woo%2C+O">Okkyun Woo</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+W">Wonhyuk Yang</a>, <a href="/search/cs?searchtype=author&query=Bae%2C+J">Jinhoon Bae</a>, <a href="/search/cs?searchtype=author&query=Park%2C+E">Eunhyeok Park</a>, <a href="/search/cs?searchtype=author&query=Sung%2C+H">Hyojin Sung</a>, <a href="/search/cs?searchtype=author&query=Lim%2C+E">Euicheol Lim</a>, <a href="/search/cs?searchtype=author&query=Kim%2C+G">Gwangsun Kim</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2404.19381v3-abstract-short" style="display: inline;"> Emerging Compute Express Link (CXL) enables cost-efficient memory expansion beyond the local DRAM of processors. While its CXL$.$mem protocol provides minimal latency overhead through an optimized protocol stack, frequent CXL memory accesses can result in significant slowdowns for memory-bound applications whether they are latency-sensitive or bandwidth-intensive. The near-data processing (NDP) in… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2404.19381v3-abstract-full').style.display = 'inline'; document.getElementById('2404.19381v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2404.19381v3-abstract-full" style="display: none;"> Emerging Compute Express Link (CXL) enables cost-efficient memory expansion beyond the local DRAM of processors. While its CXL$.$mem protocol provides minimal latency overhead through an optimized protocol stack, frequent CXL memory accesses can result in significant slowdowns for memory-bound applications whether they are latency-sensitive or bandwidth-intensive. The near-data processing (NDP) in the CXL controller promises to overcome such limitations of passive CXL memory. However, prior work on NDP in CXL memory proposes application-specific units that are not suitable for practical CXL memory-based systems that should support various applications. On the other hand, existing CPU or GPU cores are not cost-effective for NDP because they are not optimized for memory-bound applications. In addition, the communication between the host processor and CXL controller for NDP offloading should achieve low latency, but existing CXL$.$io/PCIe-based mechanisms incur $渭$s-scale latency and are not suitable for fine-grained NDP. To achieve high-performance NDP end-to-end, we propose a low-overhead general-purpose NDP architecture for CXL memory referred to as Memory-Mapped NDP (M$^2$NDP), which comprises memory-mapped functions (M$^2$func) and memory-mapped $渭$threading (M$^2渭$thread). M$^2$func is a CXL$.$mem-compatible low-overhead communication mechanism between the host processor and NDP controller in CXL memory. M$^2渭$thread enables low-cost, general-purpose NDP unit design by introducing lightweight $渭$threads that support highly concurrent execution of kernels with minimal resource wastage. Combining them, M$^2$NDP achieves significant speedups for various workloads by up to 128x (14.5x overall) and reduces energy by up to 87.9% (80.3% overall) compared to baseline CPU/GPU hosts with passive CXL memory. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2404.19381v3-abstract-full').style.display = 'none'; document.getElementById('2404.19381v3-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 23 September, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 30 April, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Accepted at the 57th IEEE/ACM International Symposium on Microarchitecture (MICRO), 2024</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2404.14243">arXiv:2404.14243</a> <span> [<a href="https://arxiv.org/pdf/2404.14243">pdf</a>, <a href="https://arxiv.org/format/2404.14243">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Information Retrieval">cs.IR</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Information Theory">cs.IT</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Social and Information Networks">cs.SI</span> </div> </div> <p class="title is-5 mathjax"> Turbo-CF: Matrix Decomposition-Free Graph Filtering for Fast Recommendation </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Park%2C+J">Jin-Duk Park</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yong-Min Shin</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+W">Won-Yong Shin</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2404.14243v1-abstract-short" style="display: inline;"> A series of graph filtering (GF)-based collaborative filtering (CF) showcases state-of-the-art performance on the recommendation accuracy by using a low-pass filter (LPF) without a training process. However, conventional GF-based CF approaches mostly perform matrix decomposition on the item-item similarity graph to realize the ideal LPF, which results in a non-trivial computational cost and thus m… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2404.14243v1-abstract-full').style.display = 'inline'; document.getElementById('2404.14243v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2404.14243v1-abstract-full" style="display: none;"> A series of graph filtering (GF)-based collaborative filtering (CF) showcases state-of-the-art performance on the recommendation accuracy by using a low-pass filter (LPF) without a training process. However, conventional GF-based CF approaches mostly perform matrix decomposition on the item-item similarity graph to realize the ideal LPF, which results in a non-trivial computational cost and thus makes them less practical in scenarios where rapid recommendations are essential. In this paper, we propose Turbo-CF, a GF-based CF method that is both training-free and matrix decomposition-free. Turbo-CF employs a polynomial graph filter to circumvent the issue of expensive matrix decompositions, enabling us to make full use of modern computer hardware components (i.e., GPU). Specifically, Turbo-CF first constructs an item-item similarity graph whose edge weights are effectively regulated. Then, our own polynomial LPFs are designed to retain only low-frequency signals without explicit matrix decompositions. We demonstrate that Turbo-CF is extremely fast yet accurate, achieving a runtime of less than 1 second on real-world benchmark datasets while achieving recommendation accuracies comparable to best competitors. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2404.14243v1-abstract-full').style.display = 'none'; document.getElementById('2404.14243v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 22 April, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">5 pages, 4 figures, 4 tables; 47th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR 2024) (to appear) (Please cite our conference version.)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2403.13831">arXiv:2403.13831</a> <span> [<a href="https://arxiv.org/pdf/2403.13831">pdf</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Emerging Technologies">cs.ET</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Human-Computer Interaction">cs.HC</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Optics">physics.optics</span> </div> </div> <p class="title is-5 mathjax"> Dual-sided transparent display </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Halder%2C+S">Suman Halder</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yunho Shin</a>, <a href="/search/cs?searchtype=author&query=Peng%2C+Y">Yidan Peng</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+L">Long Wang</a>, <a href="/search/cs?searchtype=author&query=Duan%2C+L">Liye Duan</a>, <a href="/search/cs?searchtype=author&query=Schmalenberg%2C+P">Paul Schmalenberg</a>, <a href="/search/cs?searchtype=author&query=Qin%2C+G">Guangkui Qin</a>, <a href="/search/cs?searchtype=author&query=Gao%2C+Y">Yuxi Gao</a>, <a href="/search/cs?searchtype=author&query=Dede%2C+E+M">Ercan M. Dede</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+D">Deng-Ke Yang</a>, <a href="/search/cs?searchtype=author&query=Rodrigues%2C+S+P">Sean P. Rodrigues</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2403.13831v1-abstract-short" style="display: inline;"> In the past decade, display technology has been reimagined to meet the needs of the virtual world. By mapping information onto a scene through a transparent display, users can simultaneously visualize both the real world and layers of virtual elements. However, advances in augmented reality (AR) technology have primarily focused on wearable gear or personal devices. Here we present a single displa… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2403.13831v1-abstract-full').style.display = 'inline'; document.getElementById('2403.13831v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2403.13831v1-abstract-full" style="display: none;"> In the past decade, display technology has been reimagined to meet the needs of the virtual world. By mapping information onto a scene through a transparent display, users can simultaneously visualize both the real world and layers of virtual elements. However, advances in augmented reality (AR) technology have primarily focused on wearable gear or personal devices. Here we present a single display capable of delivering visual information to observers positioned on either side of the transparent device. This dual-sided display system employs a polymer stabilized liquid crystal waveguide technology to achieve a transparency window of 65% while offering active-matrix control. An early-stage prototype exhibits full-color information via time-sequential processing of a red-green-blue (RGB) light-emitting diode (LED) strip. The dual-sided display provides a perspective on transparent mediums as display devices for human-centric and service-related experiences that can support both enhanced bi-directional user interactions and new media platforms. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2403.13831v1-abstract-full').style.display = 'none'; document.getElementById('2403.13831v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 7 March, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2403.10748">arXiv:2403.10748</a> <span> [<a href="https://arxiv.org/pdf/2403.10748">pdf</a>, <a href="https://arxiv.org/format/2403.10748">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computational Engineering, Finance, and Science">cs.CE</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Mathematical Software">cs.MS</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Numerical Analysis">math.NA</span> </div> </div> <p class="title is-5 mathjax"> A Comprehensive Review of Latent Space Dynamics Identification Algorithms for Intrusive and Non-Intrusive Reduced-Order-Modeling </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Bonneville%2C+C">Christophe Bonneville</a>, <a href="/search/cs?searchtype=author&query=He%2C+X">Xiaolong He</a>, <a href="/search/cs?searchtype=author&query=Tran%2C+A">April Tran</a>, <a href="/search/cs?searchtype=author&query=Park%2C+J+S">Jun Sur Park</a>, <a href="/search/cs?searchtype=author&query=Fries%2C+W">William Fries</a>, <a href="/search/cs?searchtype=author&query=Messenger%2C+D+A">Daniel A. Messenger</a>, <a href="/search/cs?searchtype=author&query=Cheung%2C+S+W">Siu Wun Cheung</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yeonjong Shin</a>, <a href="/search/cs?searchtype=author&query=Bortz%2C+D+M">David M. Bortz</a>, <a href="/search/cs?searchtype=author&query=Ghosh%2C+D">Debojyoti Ghosh</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+J">Jiun-Shyan Chen</a>, <a href="/search/cs?searchtype=author&query=Belof%2C+J">Jonathan Belof</a>, <a href="/search/cs?searchtype=author&query=Choi%2C+Y">Youngsoo Choi</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2403.10748v1-abstract-short" style="display: inline;"> Numerical solvers of partial differential equations (PDEs) have been widely employed for simulating physical systems. However, the computational cost remains a major bottleneck in various scientific and engineering applications, which has motivated the development of reduced-order models (ROMs). Recently, machine-learning-based ROMs have gained significant popularity and are promising for addressi… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2403.10748v1-abstract-full').style.display = 'inline'; document.getElementById('2403.10748v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2403.10748v1-abstract-full" style="display: none;"> Numerical solvers of partial differential equations (PDEs) have been widely employed for simulating physical systems. However, the computational cost remains a major bottleneck in various scientific and engineering applications, which has motivated the development of reduced-order models (ROMs). Recently, machine-learning-based ROMs have gained significant popularity and are promising for addressing some limitations of traditional ROM methods, especially for advection dominated systems. In this chapter, we focus on a particular framework known as Latent Space Dynamics Identification (LaSDI), which transforms the high-fidelity data, governed by a PDE, to simpler and low-dimensional latent-space data, governed by ordinary differential equations (ODEs). These ODEs can be learned and subsequently interpolated to make ROM predictions. Each building block of LaSDI can be easily modulated depending on the application, which makes the LaSDI framework highly flexible. In particular, we present strategies to enforce the laws of thermodynamics into LaSDI models (tLaSDI), enhance robustness in the presence of noise through the weak form (WLaSDI), select high-fidelity training data efficiently through active learning (gLaSDI, GPLaSDI), and quantify the ROM prediction uncertainty through Gaussian processes (GPLaSDI). We demonstrate the performance of different LaSDI approaches on Burgers equation, a non-linear heat conduction problem, and a plasma physics problem, showing that LaSDI algorithms can achieve relative errors of less than a few percent and up to thousands of times speed-ups. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2403.10748v1-abstract-full').style.display = 'none'; document.getElementById('2403.10748v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 15 March, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2403.05848">arXiv:2403.05848</a> <span> [<a href="https://arxiv.org/pdf/2403.05848">pdf</a>, <a href="https://arxiv.org/format/2403.05848">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Dynamical Systems">math.DS</span> </div> </div> <p class="title is-5 mathjax"> tLaSDI: Thermodynamics-informed latent space dynamics identification </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Park%2C+J+S+R">Jun Sur Richard Park</a>, <a href="/search/cs?searchtype=author&query=Cheung%2C+S+W">Siu Wun Cheung</a>, <a href="/search/cs?searchtype=author&query=Choi%2C+Y">Youngsoo Choi</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yeonjong Shin</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2403.05848v2-abstract-short" style="display: inline;"> We propose a latent space dynamics identification method, namely tLaSDI, that embeds the first and second principles of thermodynamics. The latent variables are learned through an autoencoder as a nonlinear dimension reduction model. The latent dynamics are constructed by a neural network-based model that precisely preserves certain structures for the thermodynamic laws through the GENERIC formali… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2403.05848v2-abstract-full').style.display = 'inline'; document.getElementById('2403.05848v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2403.05848v2-abstract-full" style="display: none;"> We propose a latent space dynamics identification method, namely tLaSDI, that embeds the first and second principles of thermodynamics. The latent variables are learned through an autoencoder as a nonlinear dimension reduction model. The latent dynamics are constructed by a neural network-based model that precisely preserves certain structures for the thermodynamic laws through the GENERIC formalism. An abstract error estimate is established, which provides a new loss formulation involving the Jacobian computation of autoencoder. The autoencoder and the latent dynamics are simultaneously trained to minimize the new loss. Computational examples demonstrate the effectiveness of tLaSDI, which exhibits robust generalization ability, even in extrapolation. In addition, an intriguing correlation is empirically observed between a quantity from tLaSDI in the latent space and the behaviors of the full-state solution. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2403.05848v2-abstract-full').style.display = 'none'; document.getElementById('2403.05848v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 21 March, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 9 March, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">32 pages, 8 figures</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2402.08138">arXiv:2402.08138</a> <span> [<a href="https://arxiv.org/pdf/2402.08138">pdf</a>, <a href="https://arxiv.org/format/2402.08138">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> H2O-SDF: Two-phase Learning for 3D Indoor Reconstruction using Object Surface Fields </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Park%2C+M">Minyoung Park</a>, <a href="/search/cs?searchtype=author&query=Do%2C+M">Mirae Do</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">YeonJae Shin</a>, <a href="/search/cs?searchtype=author&query=Yoo%2C+J">Jaeseok Yoo</a>, <a href="/search/cs?searchtype=author&query=Hong%2C+J">Jongkwang Hong</a>, <a href="/search/cs?searchtype=author&query=Kim%2C+J">Joongrock Kim</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+C">Chul Lee</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2402.08138v2-abstract-short" style="display: inline;"> Advanced techniques using Neural Radiance Fields (NeRF), Signed Distance Fields (SDF), and Occupancy Fields have recently emerged as solutions for 3D indoor scene reconstruction. We introduce a novel two-phase learning approach, H2O-SDF, that discriminates between object and non-object regions within indoor environments. This method achieves a nuanced balance, carefully preserving the geometric in… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2402.08138v2-abstract-full').style.display = 'inline'; document.getElementById('2402.08138v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2402.08138v2-abstract-full" style="display: none;"> Advanced techniques using Neural Radiance Fields (NeRF), Signed Distance Fields (SDF), and Occupancy Fields have recently emerged as solutions for 3D indoor scene reconstruction. We introduce a novel two-phase learning approach, H2O-SDF, that discriminates between object and non-object regions within indoor environments. This method achieves a nuanced balance, carefully preserving the geometric integrity of room layouts while also capturing intricate surface details of specific objects. A cornerstone of our two-phase learning framework is the introduction of the Object Surface Field (OSF), a novel concept designed to mitigate the persistent vanishing gradient problem that has previously hindered the capture of high-frequency details in other methods. Our proposed approach is validated through several experiments that include ablation studies. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2402.08138v2-abstract-full').style.display = 'none'; document.getElementById('2402.08138v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 8 March, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 12 February, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2401.17019">arXiv:2401.17019</a> <span> [<a href="https://arxiv.org/pdf/2401.17019">pdf</a>, <a href="https://arxiv.org/format/2401.17019">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Software Engineering">cs.SE</span> </div> </div> <p class="title is-5 mathjax"> Towards Generating Executable Metamorphic Relations Using Large Language Models </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Shin%2C+S+Y">Seung Yeob Shin</a>, <a href="/search/cs?searchtype=author&query=Pastore%2C+F">Fabrizio Pastore</a>, <a href="/search/cs?searchtype=author&query=Bianculli%2C+D">Domenico Bianculli</a>, <a href="/search/cs?searchtype=author&query=Baicoianu%2C+A">Alexandra Baicoianu</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2401.17019v3-abstract-short" style="display: inline;"> Metamorphic testing (MT) has proven to be a successful solution to automating testing and addressing the oracle problem. However, it entails manually deriving metamorphic relations (MRs) and converting them into an executable form; these steps are time-consuming and may prevent the adoption of MT. In this paper, we propose an approach for automatically deriving executable MRs (EMRs) from requireme… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2401.17019v3-abstract-full').style.display = 'inline'; document.getElementById('2401.17019v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2401.17019v3-abstract-full" style="display: none;"> Metamorphic testing (MT) has proven to be a successful solution to automating testing and addressing the oracle problem. However, it entails manually deriving metamorphic relations (MRs) and converting them into an executable form; these steps are time-consuming and may prevent the adoption of MT. In this paper, we propose an approach for automatically deriving executable MRs (EMRs) from requirements using large language models (LLMs). Instead of merely asking the LLM to produce EMRs, our approach relies on a few-shot prompting strategy to instruct the LLM to perform activities in the MT process, by providing requirements and API specifications, as one would do with software engineers. To assess the feasibility of our approach, we conducted a questionnaire-based survey in collaboration with Siemens Industry Software, a worldwide leader in providing industry software and services, focusing on four of their software applications. Additionally, we evaluated the accuracy of the generated EMRs for a Web application. The outcomes of our study are highly promising, as they demonstrate the capability of our approach to generate MRs and EMRs that are both comprehensible and pertinent for testing purposes. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2401.17019v3-abstract-full').style.display = 'none'; document.getElementById('2401.17019v3-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 11 October, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 30 January, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> January 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">This preprint has not undergone peer review (when applicable) or any post-submission improvements or corrections. The Version of Record of this contribution is published in Communications in Computer and Information Science (CCIS, volume 2178), and is available online at https://doi.org/10.1007/978-3-031-70245-7_9</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2401.03717">arXiv:2401.03717</a> <span> [<a href="https://arxiv.org/pdf/2401.03717">pdf</a>, <a href="https://arxiv.org/format/2401.03717">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Universal Time-Series Representation Learning: A Survey </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Trirat%2C+P">Patara Trirat</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yooju Shin</a>, <a href="/search/cs?searchtype=author&query=Kang%2C+J">Junhyeok Kang</a>, <a href="/search/cs?searchtype=author&query=Nam%2C+Y">Youngeun Nam</a>, <a href="/search/cs?searchtype=author&query=Na%2C+J">Jihye Na</a>, <a href="/search/cs?searchtype=author&query=Bae%2C+M">Minyoung Bae</a>, <a href="/search/cs?searchtype=author&query=Kim%2C+J">Joeun Kim</a>, <a href="/search/cs?searchtype=author&query=Kim%2C+B">Byunghyun Kim</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+J">Jae-Gil Lee</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2401.03717v3-abstract-short" style="display: inline;"> Time-series data exists in every corner of real-world systems and services, ranging from satellites in the sky to wearable devices on human bodies. Learning representations by extracting and inferring valuable information from these time series is crucial for understanding the complex dynamics of particular phenomena and enabling informed decisions. With the learned representations, we can perform… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2401.03717v3-abstract-full').style.display = 'inline'; document.getElementById('2401.03717v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2401.03717v3-abstract-full" style="display: none;"> Time-series data exists in every corner of real-world systems and services, ranging from satellites in the sky to wearable devices on human bodies. Learning representations by extracting and inferring valuable information from these time series is crucial for understanding the complex dynamics of particular phenomena and enabling informed decisions. With the learned representations, we can perform numerous downstream analyses more effectively. Among several approaches, deep learning has demonstrated remarkable performance in extracting hidden patterns and features from time-series data without manual feature engineering. This survey first presents a novel taxonomy based on three fundamental elements in designing state-of-the-art universal representation learning methods for time series. According to the proposed taxonomy, we comprehensively review existing studies and discuss their intuitions and insights into how these methods enhance the quality of learned representations. Finally, as a guideline for future studies, we summarize commonly used experimental setups and datasets and discuss several promising research directions. An up-to-date corresponding resource is available at https://github.com/itouchz/awesome-deep-time-series-representations. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2401.03717v3-abstract-full').style.display = 'none'; document.getElementById('2401.03717v3-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 27 August, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 8 January, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> January 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">41 pages, 7 figures</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2401.03676">arXiv:2401.03676</a> <span> [<a href="https://arxiv.org/pdf/2401.03676">pdf</a>, <a href="https://arxiv.org/format/2401.03676">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Software Engineering">cs.SE</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Assessing AI Detectors in Identifying AI-Generated Code: Implications for Education </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pan%2C+W+H">Wei Hung Pan</a>, <a href="/search/cs?searchtype=author&query=Chok%2C+M+J">Ming Jie Chok</a>, <a href="/search/cs?searchtype=author&query=Wong%2C+J+L+S">Jonathan Leong Shan Wong</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y+X">Yung Xin Shin</a>, <a href="/search/cs?searchtype=author&query=Poon%2C+Y+S">Yeong Shian Poon</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+Z">Zhou Yang</a>, <a href="/search/cs?searchtype=author&query=Chong%2C+C+Y">Chun Yong Chong</a>, <a href="/search/cs?searchtype=author&query=Lo%2C+D">David Lo</a>, <a href="/search/cs?searchtype=author&query=Lim%2C+M+K">Mei Kuan Lim</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2401.03676v1-abstract-short" style="display: inline;"> Educators are increasingly concerned about the usage of Large Language Models (LLMs) such as ChatGPT in programming education, particularly regarding the potential exploitation of imperfections in Artificial Intelligence Generated Content (AIGC) Detectors for academic misconduct. In this paper, we present an empirical study where the LLM is examined for its attempts to bypass detection by AIGC Det… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2401.03676v1-abstract-full').style.display = 'inline'; document.getElementById('2401.03676v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2401.03676v1-abstract-full" style="display: none;"> Educators are increasingly concerned about the usage of Large Language Models (LLMs) such as ChatGPT in programming education, particularly regarding the potential exploitation of imperfections in Artificial Intelligence Generated Content (AIGC) Detectors for academic misconduct. In this paper, we present an empirical study where the LLM is examined for its attempts to bypass detection by AIGC Detectors. This is achieved by generating code in response to a given question using different variants. We collected a dataset comprising 5,069 samples, with each sample consisting of a textual description of a coding problem and its corresponding human-written Python solution codes. These samples were obtained from various sources, including 80 from Quescol, 3,264 from Kaggle, and 1,725 from LeetCode. From the dataset, we created 13 sets of code problem variant prompts, which were used to instruct ChatGPT to generate the outputs. Subsequently, we assessed the performance of five AIGC detectors. Our results demonstrate that existing AIGC Detectors perform poorly in distinguishing between human-written code and AI-generated code. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2401.03676v1-abstract-full').style.display = 'none'; document.getElementById('2401.03676v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 8 January, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> January 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">11 pages, paper accepted at 46th International Conference on Software Engineering, Software Engineering Education and Training Track (ICSE-SEET 2024)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2312.17507">arXiv:2312.17507</a> <span> [<a href="https://arxiv.org/pdf/2312.17507">pdf</a>, <a href="https://arxiv.org/format/2312.17507">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Robotics">cs.RO</span> </div> </div> <p class="title is-5 mathjax"> Actuator-Constrained Reinforcement Learning for High-Speed Quadrupedal Locomotion </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Young-Ha Shin</a>, <a href="/search/cs?searchtype=author&query=Song%2C+T">Tae-Gyu Song</a>, <a href="/search/cs?searchtype=author&query=Ji%2C+G">Gwanghyeon Ji</a>, <a href="/search/cs?searchtype=author&query=Park%2C+H">Hae-Won Park</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2312.17507v1-abstract-short" style="display: inline;"> This paper presents a method for achieving high-speed running of a quadruped robot by considering the actuator torque-speed operating region in reinforcement learning. The physical properties and constraints of the actuator are included in the training process to reduce state transitions that are infeasible in the real world due to motor torque-speed limitations. The gait reward is designed to dis… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.17507v1-abstract-full').style.display = 'inline'; document.getElementById('2312.17507v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2312.17507v1-abstract-full" style="display: none;"> This paper presents a method for achieving high-speed running of a quadruped robot by considering the actuator torque-speed operating region in reinforcement learning. The physical properties and constraints of the actuator are included in the training process to reduce state transitions that are infeasible in the real world due to motor torque-speed limitations. The gait reward is designed to distribute motor torque evenly across all legs, contributing to more balanced power usage and mitigating performance bottlenecks due to single-motor saturation. Additionally, we designed a lightweight foot to enhance the robot's agility. We observed that applying the motor operating region as a constraint helps the policy network avoid infeasible areas during sampling. With the trained policy, KAIST Hound, a 45 kg quadruped robot, can run up to 6.5 m/s, which is the fastest speed among electric motor-based quadruped robots. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.17507v1-abstract-full').style.display = 'none'; document.getElementById('2312.17507v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 29 December, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2312.16581">arXiv:2312.16581</a> <span> [<a href="https://arxiv.org/pdf/2312.16581">pdf</a>, <a href="https://arxiv.org/format/2312.16581">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Information Retrieval">cs.IR</span> </div> </div> <p class="title is-5 mathjax"> Continuous-time Autoencoders for Regular and Irregular Time Series Imputation </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Wi%2C+H">Hyowon Wi</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yehjin Shin</a>, <a href="/search/cs?searchtype=author&query=Park%2C+N">Noseong Park</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2312.16581v3-abstract-short" style="display: inline;"> Time series imputation is one of the most fundamental tasks for time series. Real-world time series datasets are frequently incomplete (or irregular with missing observations), in which case imputation is strongly required. Many different time series imputation methods have been proposed. Recent self-attention-based methods show the state-of-the-art imputation performance. However, it has been ove… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.16581v3-abstract-full').style.display = 'inline'; document.getElementById('2312.16581v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2312.16581v3-abstract-full" style="display: none;"> Time series imputation is one of the most fundamental tasks for time series. Real-world time series datasets are frequently incomplete (or irregular with missing observations), in which case imputation is strongly required. Many different time series imputation methods have been proposed. Recent self-attention-based methods show the state-of-the-art imputation performance. However, it has been overlooked for a long time to design an imputation method based on continuous-time recurrent neural networks (RNNs), i.e., neural controlled differential equations (NCDEs). To this end, we redesign time series (variational) autoencoders based on NCDEs. Our method, called continuous-time autoencoder (CTA), encodes an input time series sample into a continuous hidden path (rather than a hidden vector) and decodes it to reconstruct and impute the input. In our experiments with 4 datasets and 19 baselines, our method shows the best imputation performance in almost all cases. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.16581v3-abstract-full').style.display = 'none'; document.getElementById('2312.16581v3-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 24 June, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 27 December, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Published as a WSDM'24 full paper (oral presentation)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2312.10325">arXiv:2312.10325</a> <span> [<a href="https://arxiv.org/pdf/2312.10325">pdf</a>, <a href="https://arxiv.org/format/2312.10325">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Information Retrieval">cs.IR</span> </div> </div> <p class="title is-5 mathjax"> An Attentive Inductive Bias for Sequential Recommendation beyond the Self-Attention </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yehjin Shin</a>, <a href="/search/cs?searchtype=author&query=Choi%2C+J">Jeongwhan Choi</a>, <a href="/search/cs?searchtype=author&query=Wi%2C+H">Hyowon Wi</a>, <a href="/search/cs?searchtype=author&query=Park%2C+N">Noseong Park</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2312.10325v2-abstract-short" style="display: inline;"> Sequential recommendation (SR) models based on Transformers have achieved remarkable successes. The self-attention mechanism of Transformers for computer vision and natural language processing suffers from the oversmoothing problem, i.e., hidden representations becoming similar to tokens. In the SR domain, we, for the first time, show that the same problem occurs. We present pioneering investigati… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.10325v2-abstract-full').style.display = 'inline'; document.getElementById('2312.10325v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2312.10325v2-abstract-full" style="display: none;"> Sequential recommendation (SR) models based on Transformers have achieved remarkable successes. The self-attention mechanism of Transformers for computer vision and natural language processing suffers from the oversmoothing problem, i.e., hidden representations becoming similar to tokens. In the SR domain, we, for the first time, show that the same problem occurs. We present pioneering investigations that reveal the low-pass filtering nature of self-attention in the SR, which causes oversmoothing. To this end, we propose a novel method called $\textbf{B}$eyond $\textbf{S}$elf-$\textbf{A}$ttention for Sequential $\textbf{Rec}$ommendation (BSARec), which leverages the Fourier transform to i) inject an inductive bias by considering fine-grained sequential patterns and ii) integrate low and high-frequency information to mitigate oversmoothing. Our discovery shows significant advancements in the SR domain and is expected to bridge the gap for existing Transformer-based SR models. We test our proposed approach through extensive experiments on 6 benchmark datasets. The experimental results demonstrate that our model outperforms 7 baseline methods in terms of recommendation performance. Our code is available at https://github.com/yehjin-shin/BSARec. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.10325v2-abstract-full').style.display = 'none'; document.getElementById('2312.10325v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 17 February, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 16 December, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Accepted by AAAI 2024. Yehjin Shin and Jeongwhan Choi are co-first authors with equal contribution</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2312.10072">arXiv:2312.10072</a> <span> [<a href="https://arxiv.org/pdf/2312.10072">pdf</a>, <a href="https://arxiv.org/format/2312.10072">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Human-Computer Interaction">cs.HC</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Applications">stat.AP</span> </div> </div> <p class="title is-5 mathjax"> Assessing the Usability of GutGPT: A Simulation Study of an AI Clinical Decision Support System for Gastrointestinal Bleeding Risk </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chan%2C+C">Colleen Chan</a>, <a href="/search/cs?searchtype=author&query=You%2C+K">Kisung You</a>, <a href="/search/cs?searchtype=author&query=Chung%2C+S">Sunny Chung</a>, <a href="/search/cs?searchtype=author&query=Giuffr%C3%A8%2C+M">Mauro Giuffr猫</a>, <a href="/search/cs?searchtype=author&query=Saarinen%2C+T">Theo Saarinen</a>, <a href="/search/cs?searchtype=author&query=Rajashekar%2C+N">Niroop Rajashekar</a>, <a href="/search/cs?searchtype=author&query=Pu%2C+Y">Yuan Pu</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y+E">Yeo Eun Shin</a>, <a href="/search/cs?searchtype=author&query=Laine%2C+L">Loren Laine</a>, <a href="/search/cs?searchtype=author&query=Wong%2C+A">Ambrose Wong</a>, <a href="/search/cs?searchtype=author&query=Kizilcec%2C+R">Ren茅 Kizilcec</a>, <a href="/search/cs?searchtype=author&query=Sekhon%2C+J">Jasjeet Sekhon</a>, <a href="/search/cs?searchtype=author&query=Shung%2C+D">Dennis Shung</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2312.10072v1-abstract-short" style="display: inline;"> Applications of large language models (LLMs) like ChatGPT have potential to enhance clinical decision support through conversational interfaces. However, challenges of human-algorithmic interaction and clinician trust are poorly understood. GutGPT, a LLM for gastrointestinal (GI) bleeding risk prediction and management guidance, was deployed in clinical simulation scenarios alongside the electroni… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.10072v1-abstract-full').style.display = 'inline'; document.getElementById('2312.10072v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2312.10072v1-abstract-full" style="display: none;"> Applications of large language models (LLMs) like ChatGPT have potential to enhance clinical decision support through conversational interfaces. However, challenges of human-algorithmic interaction and clinician trust are poorly understood. GutGPT, a LLM for gastrointestinal (GI) bleeding risk prediction and management guidance, was deployed in clinical simulation scenarios alongside the electronic health record (EHR) with emergency medicine physicians, internal medicine physicians, and medical students to evaluate its effect on physician acceptance and trust in AI clinical decision support systems (AI-CDSS). GutGPT provides risk predictions from a validated machine learning model and evidence-based answers by querying extracted clinical guidelines. Participants were randomized to GutGPT and an interactive dashboard, or the interactive dashboard and a search engine. Surveys and educational assessments taken before and after measured technology acceptance and content mastery. Preliminary results showed mixed effects on acceptance after using GutGPT compared to the dashboard or search engine but appeared to improve content mastery based on simulation performance. Overall, this study demonstrates LLMs like GutGPT could enhance effective AI-CDSS if implemented optimally and paired with interactive interfaces. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.10072v1-abstract-full').style.display = 'none'; document.getElementById('2312.10072v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 6 December, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Extended Abstract presented at Machine Learning for Health (ML4H) symposium 2023, December 10, 2023, New Orleans, United States, 11 pages</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2312.09572">arXiv:2312.09572</a> <span> [<a href="https://arxiv.org/pdf/2312.09572">pdf</a>, <a href="https://arxiv.org/format/2312.09572">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Audio and Speech Processing">eess.AS</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computation and Language">cs.CL</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Sound">cs.SD</span> </div> <div class="is-inline-block" style="margin-left: 0.5rem"> <div class="tags has-addons"> <span class="tag is-dark is-size-7">doi</span> <span class="tag is-light is-size-7"><a class="" href="https://doi.org/10.1109/ACCESS.2023.3344177">10.1109/ACCESS.2023.3344177 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> IR-UWB Radar-Based Contactless Silent Speech Recognition of Vowels, Consonants, Words, and Phrases </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lee%2C+S">Sunghwa Lee</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Younghoon Shin</a>, <a href="/search/cs?searchtype=author&query=Kim%2C+M">Myungjong Kim</a>, <a href="/search/cs?searchtype=author&query=Seo%2C+J">Jiwon Seo</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2312.09572v1-abstract-short" style="display: inline;"> Several sensing techniques have been proposed for silent speech recognition (SSR); however, many of these methods require invasive processes or sensor attachment to the skin using adhesive tape or glue, rendering them unsuitable for frequent use in daily life. By contrast, impulse radio ultra-wideband (IR-UWB) radar can operate without physical contact with users' articulators and related body par… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.09572v1-abstract-full').style.display = 'inline'; document.getElementById('2312.09572v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2312.09572v1-abstract-full" style="display: none;"> Several sensing techniques have been proposed for silent speech recognition (SSR); however, many of these methods require invasive processes or sensor attachment to the skin using adhesive tape or glue, rendering them unsuitable for frequent use in daily life. By contrast, impulse radio ultra-wideband (IR-UWB) radar can operate without physical contact with users' articulators and related body parts, offering several advantages for SSR. These advantages include high range resolution, high penetrability, low power consumption, robustness to external light or sound interference, and the ability to be embedded in space-constrained handheld devices. This study demonstrated IR-UWB radar-based contactless SSR using four types of speech stimuli (vowels, consonants, words, and phrases). To achieve this, a novel speech feature extraction algorithm specifically designed for IR-UWB radar-based SSR is proposed. Each speech stimulus is recognized by applying a classification algorithm to the extracted speech features. Two different algorithms, multidimensional dynamic time warping (MD-DTW) and deep neural network-hidden Markov model (DNN-HMM), were compared for the classification task. Additionally, a favorable radar antenna position, either in front of the user's lips or below the user's chin, was determined to achieve higher recognition accuracy. Experimental results demonstrated the efficacy of the proposed speech feature extraction algorithm combined with DNN-HMM for classifying vowels, consonants, words, and phrases. Notably, this study represents the first demonstration of phoneme-level SSR using contactless radar. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.09572v1-abstract-full').style.display = 'none'; document.getElementById('2312.09572v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 15 December, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Submitted to IEEE Access</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2312.08677">arXiv:2312.08677</a> <span> [<a href="https://arxiv.org/pdf/2312.08677">pdf</a>, <a href="https://arxiv.org/format/2312.08677">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Adaptive Shortcut Debiasing for Online Continual Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Kim%2C+D">Doyoung Kim</a>, <a href="/search/cs?searchtype=author&query=Park%2C+D">Dongmin Park</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yooju Shin</a>, <a href="/search/cs?searchtype=author&query=Bang%2C+J">Jihwan Bang</a>, <a href="/search/cs?searchtype=author&query=Song%2C+H">Hwanjun Song</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+J">Jae-Gil Lee</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2312.08677v1-abstract-short" style="display: inline;"> We propose a novel framework DropTop that suppresses the shortcut bias in online continual learning (OCL) while being adaptive to the varying degree of the shortcut bias incurred by continuously changing environment. By the observed high-attention property of the shortcut bias, highly-activated features are considered candidates for debiasing. More importantly, resolving the limitation of the onli… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.08677v1-abstract-full').style.display = 'inline'; document.getElementById('2312.08677v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2312.08677v1-abstract-full" style="display: none;"> We propose a novel framework DropTop that suppresses the shortcut bias in online continual learning (OCL) while being adaptive to the varying degree of the shortcut bias incurred by continuously changing environment. By the observed high-attention property of the shortcut bias, highly-activated features are considered candidates for debiasing. More importantly, resolving the limitation of the online environment where prior knowledge and auxiliary data are not ready, two novel techniques -- feature map fusion and adaptive intensity shifting -- enable us to automatically determine the appropriate level and proportion of the candidate shortcut features to be dropped. Extensive experiments on five benchmark datasets demonstrate that, when combined with various OCL algorithms, DropTop increases the average accuracy by up to 10.4% and decreases the forgetting by up to 63.2%. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.08677v1-abstract-full').style.display = 'none'; document.getElementById('2312.08677v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 14 December, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2312.07753">arXiv:2312.07753</a> <span> [<a href="https://arxiv.org/pdf/2312.07753">pdf</a>, <a href="https://arxiv.org/format/2312.07753">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Polynomial-based Self-Attention for Table Representation learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Kim%2C+J">Jayoung Kim</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yehjin Shin</a>, <a href="/search/cs?searchtype=author&query=Choi%2C+J">Jeongwhan Choi</a>, <a href="/search/cs?searchtype=author&query=Wi%2C+H">Hyowon Wi</a>, <a href="/search/cs?searchtype=author&query=Park%2C+N">Noseong Park</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2312.07753v2-abstract-short" style="display: inline;"> Structured data, which constitutes a significant portion of existing data types, has been a long-standing research topic in the field of machine learning. Various representation learning methods for tabular data have been proposed, ranging from encoder-decoder structures to Transformers. Among these, Transformer-based methods have achieved state-of-the-art performance not only in tabular data but… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.07753v2-abstract-full').style.display = 'inline'; document.getElementById('2312.07753v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2312.07753v2-abstract-full" style="display: none;"> Structured data, which constitutes a significant portion of existing data types, has been a long-standing research topic in the field of machine learning. Various representation learning methods for tabular data have been proposed, ranging from encoder-decoder structures to Transformers. Among these, Transformer-based methods have achieved state-of-the-art performance not only in tabular data but also in various other fields, including computer vision and natural language processing. However, recent studies have revealed that self-attention, a key component of Transformers, can lead to an oversmoothing issue. We show that Transformers for tabular data also face this problem, and to address the problem, we propose a novel matrix polynomial-based self-attention layer as a substitute for the original self-attention layer, which enhances model scalability. In our experiments with three representative table learning models equipped with our proposed layer, we illustrate that the layer effectively mitigates the oversmoothing problem and enhances the representation performance of the existing methods, outperforming the state-of-the-art table representation methods. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.07753v2-abstract-full').style.display = 'none'; document.getElementById('2312.07753v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 18 December, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 12 December, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2312.04234">arXiv:2312.04234</a> <span> [<a href="https://arxiv.org/pdf/2312.04234">pdf</a>, <a href="https://arxiv.org/format/2312.04234">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Graph Convolutions Enrich the Self-Attention in Transformers! </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Choi%2C+J">Jeongwhan Choi</a>, <a href="/search/cs?searchtype=author&query=Wi%2C+H">Hyowon Wi</a>, <a href="/search/cs?searchtype=author&query=Kim%2C+J">Jayoung Kim</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yehjin Shin</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+K">Kookjin Lee</a>, <a href="/search/cs?searchtype=author&query=Trask%2C+N">Nathaniel Trask</a>, <a href="/search/cs?searchtype=author&query=Park%2C+N">Noseong Park</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2312.04234v5-abstract-short" style="display: inline;"> Transformers, renowned for their self-attention mechanism, have achieved state-of-the-art performance across various tasks in natural language processing, computer vision, time-series modeling, etc. However, one of the challenges with deep Transformer models is the oversmoothing problem, where representations across layers converge to indistinguishable values, leading to significant performance de… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.04234v5-abstract-full').style.display = 'inline'; document.getElementById('2312.04234v5-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2312.04234v5-abstract-full" style="display: none;"> Transformers, renowned for their self-attention mechanism, have achieved state-of-the-art performance across various tasks in natural language processing, computer vision, time-series modeling, etc. However, one of the challenges with deep Transformer models is the oversmoothing problem, where representations across layers converge to indistinguishable values, leading to significant performance degradation. We interpret the original self-attention as a simple graph filter and redesign it from a graph signal processing (GSP) perspective. We propose a graph-filter-based self-attention (GFSA) to learn a general yet effective one, whose complexity, however, is slightly larger than that of the original self-attention mechanism. We demonstrate that GFSA improves the performance of Transformers in various fields, including computer vision, natural language processing, graph-level tasks, speech recognition, and code classification. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.04234v5-abstract-full').style.display = 'none'; document.getElementById('2312.04234v5-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 1 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 7 December, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Accepted to NeurIPS 2024. Jeongwhan Choi and Hyowon Wi are co-first authors with equal contributions</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2312.02547">arXiv:2312.02547</a> <span> [<a href="https://arxiv.org/pdf/2312.02547">pdf</a>, <a href="https://arxiv.org/format/2312.02547">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Data Structures and Algorithms">cs.DS</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computer Science and Game Theory">cs.GT</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> </div> </div> <p class="title is-5 mathjax"> On Optimal Consistency-Robustness Trade-Off for Learning-Augmented Multi-Option Ski Rental </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yongho Shin</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+C">Changyeol Lee</a>, <a href="/search/cs?searchtype=author&query=An%2C+H">Hyung-Chan An</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2312.02547v1-abstract-short" style="display: inline;"> The learning-augmented multi-option ski rental problem generalizes the classical ski rental problem in two ways: the algorithm is provided with a prediction on the number of days we can ski, and the ski rental options now come with a variety of rental periods and prices to choose from, unlike the classical two-option setting. Subsequent to the initial study of the multi-option ski rental problem (… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.02547v1-abstract-full').style.display = 'inline'; document.getElementById('2312.02547v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2312.02547v1-abstract-full" style="display: none;"> The learning-augmented multi-option ski rental problem generalizes the classical ski rental problem in two ways: the algorithm is provided with a prediction on the number of days we can ski, and the ski rental options now come with a variety of rental periods and prices to choose from, unlike the classical two-option setting. Subsequent to the initial study of the multi-option ski rental problem (without learning augmentation) due to Zhang, Poon, and Xu, significant progress has been made for this problem recently in particular. The problem is very well understood when we relinquish one of the two generalizations -- for the learning-augmented classical ski rental problem, algorithms giving best-possible trade-off between consistency and robustness exist; for the multi-option ski rental problem without learning augmentation, deterministic/randomized algorithms giving the best-possible competitiveness have been found. However, in presence of both generalizations, there remained a huge gap between the algorithmic and impossibility results. In fact, for randomized algorithms, we did not have any nontrivial lower bounds on the consistency-robustness trade-off before. This paper bridges this gap for both deterministic and randomized algorithms. For deterministic algorithms, we present a best-possible algorithm that completely matches the known lower bound. For randomized algorithms, we show the first nontrivial lower bound on the consistency-robustness trade-off, and also present an improved randomized algorithm. Our algorithm matches our lower bound on robustness within a factor of e/2 when the consistency is at most 1.086. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.02547v1-abstract-full').style.display = 'none'; document.getElementById('2312.02547v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 5 December, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">16 pages, 2 figures</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">MSC Class:</span> 68W27; 68T05 <span class="has-text-black-bis has-text-weight-semibold">ACM Class:</span> F.2.2; I.2.6 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2311.17781">arXiv:2311.17781</a> <span> [<a href="https://arxiv.org/pdf/2311.17781">pdf</a>, <a href="https://arxiv.org/format/2311.17781">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Information Theory">cs.IT</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Neural and Evolutionary Computing">cs.NE</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Social and Information Networks">cs.SI</span> </div> </div> <p class="title is-5 mathjax"> Propagate & Distill: Towards Effective Graph Learners Using Propagation-Embracing MLPs </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yong-Min Shin</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+W">Won-Yong Shin</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2311.17781v1-abstract-short" style="display: inline;"> Recent studies attempted to utilize multilayer perceptrons (MLPs) to solve semisupervised node classification on graphs, by training a student MLP by knowledge distillation from a teacher graph neural network (GNN). While previous studies have focused mostly on training the student MLP by matching the output probability distributions between the teacher and student models during distillation, it h… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2311.17781v1-abstract-full').style.display = 'inline'; document.getElementById('2311.17781v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2311.17781v1-abstract-full" style="display: none;"> Recent studies attempted to utilize multilayer perceptrons (MLPs) to solve semisupervised node classification on graphs, by training a student MLP by knowledge distillation from a teacher graph neural network (GNN). While previous studies have focused mostly on training the student MLP by matching the output probability distributions between the teacher and student models during distillation, it has not been systematically studied how to inject the structural information in an explicit and interpretable manner. Inspired by GNNs that separate feature transformation $T$ and propagation $螤$, we re-frame the distillation process as making the student MLP learn both $T$ and $螤$. Although this can be achieved by applying the inverse propagation $螤^{-1}$ before distillation from the teacher, it still comes with a high computational cost from large matrix multiplications during training. To solve this problem, we propose Propagate & Distill (P&D), which propagates the output of the teacher before distillation, which can be interpreted as an approximate process of the inverse propagation. We demonstrate that P&D can readily improve the performance of the student MLP. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2311.17781v1-abstract-full').style.display = 'none'; document.getElementById('2311.17781v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 29 November, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">17 pages, 2 figures, 8 tables; 2nd Learning on Graphs Conference (LoG 2023) (Please cite our conference version.). arXiv admin note: substantial text overlap with arXiv:2311.11759</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2311.11759">arXiv:2311.11759</a> <span> [<a href="https://arxiv.org/pdf/2311.11759">pdf</a>, <a href="https://arxiv.org/format/2311.11759">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Information Theory">cs.IT</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Neural and Evolutionary Computing">cs.NE</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Social and Information Networks">cs.SI</span> </div> </div> <p class="title is-5 mathjax"> Unveiling the Unseen Potential of Graph Learning through MLPs: Effective Graph Learners Using Propagation-Embracing MLPs </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yong-Min Shin</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+W">Won-Yong Shin</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2311.11759v1-abstract-short" style="display: inline;"> Recent studies attempted to utilize multilayer perceptrons (MLPs) to solve semi-supervised node classification on graphs, by training a student MLP by knowledge distillation (KD) from a teacher graph neural network (GNN). While previous studies have focused mostly on training the student MLP by matching the output probability distributions between the teacher and student models during KD, it has n… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2311.11759v1-abstract-full').style.display = 'inline'; document.getElementById('2311.11759v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2311.11759v1-abstract-full" style="display: none;"> Recent studies attempted to utilize multilayer perceptrons (MLPs) to solve semi-supervised node classification on graphs, by training a student MLP by knowledge distillation (KD) from a teacher graph neural network (GNN). While previous studies have focused mostly on training the student MLP by matching the output probability distributions between the teacher and student models during KD, it has not been systematically studied how to inject the structural information in an explicit and interpretable manner. Inspired by GNNs that separate feature transformation $T$ and propagation $螤$, we re-frame the KD process as enabling the student MLP to explicitly learn both $T$ and $螤$. Although this can be achieved by applying the inverse propagation $螤^{-1}$ before distillation from the teacher GNN, it still comes with a high computational cost from large matrix multiplications during training. To solve this problem, we propose Propagate & Distill (P&D), which propagates the output of the teacher GNN before KD and can be interpreted as an approximate process of the inverse propagation $螤^{-1}$. Through comprehensive evaluations using real-world benchmark datasets, we demonstrate the effectiveness of P&D by showing further performance boost of the student MLP. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2311.11759v1-abstract-full').style.display = 'none'; document.getElementById('2311.11759v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 20 November, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">35 pages, 5 figures, 8 tables</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2310.20287">arXiv:2310.20287</a> <span> [<a href="https://arxiv.org/pdf/2310.20287">pdf</a>, <a href="https://arxiv.org/format/2310.20287">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Sample-Efficient and Safe Deep Reinforcement Learning via Reset Deep Ensemble Agents </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Kim%2C+W">Woojun Kim</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yongjae Shin</a>, <a href="/search/cs?searchtype=author&query=Park%2C+J">Jongeui Park</a>, <a href="/search/cs?searchtype=author&query=Sung%2C+Y">Youngchul Sung</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2310.20287v1-abstract-short" style="display: inline;"> Deep reinforcement learning (RL) has achieved remarkable success in solving complex tasks through its integration with deep neural networks (DNNs) as function approximators. However, the reliance on DNNs has introduced a new challenge called primacy bias, whereby these function approximators tend to prioritize early experiences, leading to overfitting. To mitigate this primacy bias, a reset method… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.20287v1-abstract-full').style.display = 'inline'; document.getElementById('2310.20287v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2310.20287v1-abstract-full" style="display: none;"> Deep reinforcement learning (RL) has achieved remarkable success in solving complex tasks through its integration with deep neural networks (DNNs) as function approximators. However, the reliance on DNNs has introduced a new challenge called primacy bias, whereby these function approximators tend to prioritize early experiences, leading to overfitting. To mitigate this primacy bias, a reset method has been proposed, which performs periodic resets of a portion or the entirety of a deep RL agent while preserving the replay buffer. However, the use of the reset method can result in performance collapses after executing the reset, which can be detrimental from the perspective of safe RL and regret minimization. In this paper, we propose a new reset-based method that leverages deep ensemble learning to address the limitations of the vanilla reset method and enhance sample efficiency. The proposed method is evaluated through various experiments including those in the domain of safe RL. Numerical results show its effectiveness in high sample efficiency and safety considerations. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.20287v1-abstract-full').style.display = 'none'; document.getElementById('2310.20287v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 31 October, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">NeurIPS 2023 camera-ready</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2310.14168">arXiv:2310.14168</a> <span> [<a href="https://arxiv.org/pdf/2310.14168">pdf</a>, <a href="https://arxiv.org/format/2310.14168">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Optimization and Control">math.OC</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> </div> </div> <p class="title is-5 mathjax"> Randomized Forward Mode of Automatic Differentiation For Optimization Algorithms </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Shukla%2C+K">Khemraj Shukla</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yeonjong Shin</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2310.14168v3-abstract-short" style="display: inline;"> We present a randomized forward mode gradient (RFG) as an alternative to backpropagation. RFG is a random estimator for the gradient that is constructed based on the directional derivative along a random vector. The forward mode automatic differentiation (AD) provides an efficient computation of RFG. The probability distribution of the random vector determines the statistical properties of RFG. Th… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.14168v3-abstract-full').style.display = 'inline'; document.getElementById('2310.14168v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2310.14168v3-abstract-full" style="display: none;"> We present a randomized forward mode gradient (RFG) as an alternative to backpropagation. RFG is a random estimator for the gradient that is constructed based on the directional derivative along a random vector. The forward mode automatic differentiation (AD) provides an efficient computation of RFG. The probability distribution of the random vector determines the statistical properties of RFG. Through the second moment analysis, we found that the distribution with the smallest kurtosis yields the smallest expected relative squared error. By replacing gradient with RFG, a class of RFG-based optimization algorithms is obtained. By focusing on gradient descent (GD) and Polyak's heavy ball (PHB) methods, we present a convergence analysis of RFG-based optimization algorithms for quadratic functions. Computational experiments are presented to demonstrate the performance of the proposed algorithms and verify the theoretical findings. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.14168v3-abstract-full').style.display = 'none'; document.getElementById('2310.14168v3-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 1 February, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 22 October, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">22 Pages, 7 Figures</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">MSC Class:</span> 65K05; 65B99; 65Y20 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2310.12409">arXiv:2310.12409</a> <span> [<a href="https://arxiv.org/pdf/2310.12409">pdf</a>, <a href="https://arxiv.org/format/2310.12409">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Robotics">cs.RO</span> </div> </div> <p class="title is-5 mathjax"> Object-Aware Impedance Control for Human-Robot Collaborative Task with Online Object Parameter Estimation </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Park%2C+J">Jinseong Park</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yong-Sik Shin</a>, <a href="/search/cs?searchtype=author&query=Kim%2C+S">Sanghyun Kim</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2310.12409v1-abstract-short" style="display: inline;"> Physical human-robot interactions (pHRIs) can improve robot autonomy and reduce physical demands on humans. In this paper, we consider a collaborative task with a considerably long object and no prior knowledge of the object's parameters. An integrated control framework with an online object parameter estimator and a Cartesian object-aware impedance controller is proposed to realize complicated sc… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.12409v1-abstract-full').style.display = 'inline'; document.getElementById('2310.12409v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2310.12409v1-abstract-full" style="display: none;"> Physical human-robot interactions (pHRIs) can improve robot autonomy and reduce physical demands on humans. In this paper, we consider a collaborative task with a considerably long object and no prior knowledge of the object's parameters. An integrated control framework with an online object parameter estimator and a Cartesian object-aware impedance controller is proposed to realize complicated scenarios. During the transportation task, the object parameters are estimated online while a robot and human lift an object. The perturbation motion is incorporated into the null space of the desired trajectory to enhance the estimator accuracy. An object-aware impedance controller is designed using the real-time estimation results to effectively transmit the intended human motion to the robot through the object. Experimental demonstrations of collaborative tasks, including object transportation and assembly tasks, are implemented to show the effectiveness of our proposed method. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.12409v1-abstract-full').style.display = 'none'; document.getElementById('2310.12409v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 18 October, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">11 pages, 5 figures, for associated video, see https://youtu.be/bGH6GAFlRgA?si=wXj_SRzEE8BYoV2a</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2310.08598">arXiv:2310.08598</a> <span> [<a href="https://arxiv.org/pdf/2310.08598">pdf</a>, <a href="https://arxiv.org/format/2310.08598">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Image and Video Processing">eess.IV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> Domain Generalization for Medical Image Analysis: A Survey </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Yoon%2C+J+S">Jee Seok Yoon</a>, <a href="/search/cs?searchtype=author&query=Oh%2C+K">Kwanseok Oh</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yooseung Shin</a>, <a href="/search/cs?searchtype=author&query=Mazurowski%2C+M+A">Maciej A. Mazurowski</a>, <a href="/search/cs?searchtype=author&query=Suk%2C+H">Heung-Il Suk</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2310.08598v2-abstract-short" style="display: inline;"> Medical image analysis (MedIA) has become an essential tool in medicine and healthcare, aiding in disease diagnosis, prognosis, and treatment planning, and recent successes in deep learning (DL) have made significant contributions to its advances. However, deploying DL models for MedIA in real-world situations remains challenging due to their failure to generalize across the distributional gap bet… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.08598v2-abstract-full').style.display = 'inline'; document.getElementById('2310.08598v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2310.08598v2-abstract-full" style="display: none;"> Medical image analysis (MedIA) has become an essential tool in medicine and healthcare, aiding in disease diagnosis, prognosis, and treatment planning, and recent successes in deep learning (DL) have made significant contributions to its advances. However, deploying DL models for MedIA in real-world situations remains challenging due to their failure to generalize across the distributional gap between training and testing samples - a problem known as domain shift. Researchers have dedicated their efforts to developing various DL methods to adapt and perform robustly on unknown and out-of-distribution data distributions. This paper comprehensively reviews domain generalization studies specifically tailored for MedIA. We provide a holistic view of how domain generalization techniques interact within the broader MedIA system, going beyond methodologies to consider the operational implications on the entire MedIA workflow. Specifically, we categorize domain generalization methods into data-level, feature-level, model-level, and analysis-level methods. We show how those methods can be used in various stages of the MedIA workflow with DL equipped from data acquisition to model prediction and analysis. Furthermore, we critically analyze the strengths and weaknesses of various methods, unveiling future research opportunities. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.08598v2-abstract-full').style.display = 'none'; document.getElementById('2310.08598v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 15 February, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 5 October, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2309.01020">arXiv:2309.01020</a> <span> [<a href="https://arxiv.org/pdf/2309.01020">pdf</a>, <a href="https://arxiv.org/format/2309.01020">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Numerical Analysis">math.NA</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">stat.ML</span> </div> </div> <p class="title is-5 mathjax"> On the training and generalization of deep operator networks </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lee%2C+S">Sanghyun Lee</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yeonjong Shin</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2309.01020v1-abstract-short" style="display: inline;"> We present a novel training method for deep operator networks (DeepONets), one of the most popular neural network models for operators. DeepONets are constructed by two sub-networks, namely the branch and trunk networks. Typically, the two sub-networks are trained simultaneously, which amounts to solving a complex optimization problem in a high dimensional space. In addition, the nonconvex and non… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2309.01020v1-abstract-full').style.display = 'inline'; document.getElementById('2309.01020v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2309.01020v1-abstract-full" style="display: none;"> We present a novel training method for deep operator networks (DeepONets), one of the most popular neural network models for operators. DeepONets are constructed by two sub-networks, namely the branch and trunk networks. Typically, the two sub-networks are trained simultaneously, which amounts to solving a complex optimization problem in a high dimensional space. In addition, the nonconvex and nonlinear nature makes training very challenging. To tackle such a challenge, we propose a two-step training method that trains the trunk network first and then sequentially trains the branch network. The core mechanism is motivated by the divide-and-conquer paradigm and is the decomposition of the entire complex training task into two subtasks with reduced complexity. Therein the Gram-Schmidt orthonormalization process is introduced which significantly improves stability and generalization ability. On the theoretical side, we establish a generalization error estimate in terms of the number of training data, the width of DeepONets, and the number of input and output sensors. Numerical examples are presented to demonstrate the effectiveness of the two-step training method, including Darcy flow in heterogeneous porous media. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2309.01020v1-abstract-full').style.display = 'none'; document.getElementById('2309.01020v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 2 September, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2308.13564">arXiv:2308.13564</a> <span> [<a href="https://arxiv.org/pdf/2308.13564">pdf</a>, <a href="https://arxiv.org/format/2308.13564">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Econometrics">econ.EM</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Statistics Theory">math.ST</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computation">stat.CO</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">stat.ML</span> </div> </div> <p class="title is-5 mathjax"> SGMM: Stochastic Approximation to Generalized Method of Moments </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chen%2C+X">Xiaohong Chen</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+S">Sokbae Lee</a>, <a href="/search/cs?searchtype=author&query=Liao%2C+Y">Yuan Liao</a>, <a href="/search/cs?searchtype=author&query=Seo%2C+M+H">Myung Hwan Seo</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Youngki Shin</a>, <a href="/search/cs?searchtype=author&query=Song%2C+M">Myunghyun Song</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2308.13564v2-abstract-short" style="display: inline;"> We introduce a new class of algorithms, Stochastic Generalized Method of Moments (SGMM), for estimation and inference on (overidentified) moment restriction models. Our SGMM is a novel stochastic approximation alternative to the popular Hansen (1982) (offline) GMM, and offers fast and scalable implementation with the ability to handle streaming datasets in real time. We establish the almost sure c… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2308.13564v2-abstract-full').style.display = 'inline'; document.getElementById('2308.13564v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2308.13564v2-abstract-full" style="display: none;"> We introduce a new class of algorithms, Stochastic Generalized Method of Moments (SGMM), for estimation and inference on (overidentified) moment restriction models. Our SGMM is a novel stochastic approximation alternative to the popular Hansen (1982) (offline) GMM, and offers fast and scalable implementation with the ability to handle streaming datasets in real time. We establish the almost sure convergence, and the (functional) central limit theorem for the inefficient online 2SLS and the efficient SGMM. Moreover, we propose online versions of the Durbin-Wu-Hausman and Sargan-Hansen tests that can be seamlessly integrated within the SGMM framework. Extensive Monte Carlo simulations show that as the sample size increases, the SGMM matches the standard (offline) GMM in terms of estimation accuracy and gains over computational efficiency, indicating its practical value for both large-scale and online datasets. We demonstrate the efficacy of our approach by a proof of concept using two well known empirical examples with large sample sizes. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2308.13564v2-abstract-full').style.display = 'none'; document.getElementById('2308.13564v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 30 October, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 24 August, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> August 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">46 pages, 4 tables, 2 figures</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2308.11901">arXiv:2308.11901</a> <span> [<a href="https://arxiv.org/pdf/2308.11901">pdf</a>, <a href="https://arxiv.org/format/2308.11901">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> Camera-Driven Representation Learning for Unsupervised Domain Adaptive Person Re-identification </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lee%2C+G">Geon Lee</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+S">Sanghoon Lee</a>, <a href="/search/cs?searchtype=author&query=Kim%2C+D">Dohyung Kim</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Younghoon Shin</a>, <a href="/search/cs?searchtype=author&query=Yoon%2C+Y">Yongsang Yoon</a>, <a href="/search/cs?searchtype=author&query=Ham%2C+B">Bumsub Ham</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2308.11901v1-abstract-short" style="display: inline;"> We present a novel unsupervised domain adaption method for person re-identification (reID) that generalizes a model trained on a labeled source domain to an unlabeled target domain. We introduce a camera-driven curriculum learning (CaCL) framework that leverages camera labels of person images to transfer knowledge from source to target domains progressively. To this end, we divide target domain da… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2308.11901v1-abstract-full').style.display = 'inline'; document.getElementById('2308.11901v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2308.11901v1-abstract-full" style="display: none;"> We present a novel unsupervised domain adaption method for person re-identification (reID) that generalizes a model trained on a labeled source domain to an unlabeled target domain. We introduce a camera-driven curriculum learning (CaCL) framework that leverages camera labels of person images to transfer knowledge from source to target domains progressively. To this end, we divide target domain dataset into multiple subsets based on the camera labels, and initially train our model with a single subset (i.e., images captured by a single camera). We then gradually exploit more subsets for training, according to a curriculum sequence obtained with a camera-driven scheduling rule. The scheduler considers maximum mean discrepancies (MMD) between each subset and the source domain dataset, such that the subset closer to the source domain is exploited earlier within the curriculum. For each curriculum sequence, we generate pseudo labels of person images in a target domain to train a reID model in a supervised way. We have observed that the pseudo labels are highly biased toward cameras, suggesting that person images obtained from the same camera are likely to have the same pseudo labels, even for different IDs. To address the camera bias problem, we also introduce a camera-diversity (CD) loss encouraging person images of the same pseudo label, but captured across various cameras, to involve more for discriminative feature learning, providing person representations robust to inter-camera variations. Experimental results on standard benchmarks, including real-to-real and synthetic-to-real scenarios, demonstrate the effectiveness of our framework. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2308.11901v1-abstract-full').style.display = 'none'; document.getElementById('2308.11901v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 23 August, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> August 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Accepted to ICCV 2023</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2307.05804">arXiv:2307.05804</a> <span> [<a href="https://arxiv.org/pdf/2307.05804">pdf</a>, <a href="https://arxiv.org/format/2307.05804">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Image and Video Processing">eess.IV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> <div class="is-inline-block" style="margin-left: 0.5rem"> <div class="tags has-addons"> <span class="tag is-dark is-size-7">doi</span> <span class="tag is-light is-size-7"><a class="" href="https://doi.org/10.1016/j.compmedimag.2023.102259">10.1016/j.compmedimag.2023.102259 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Improving Segmentation and Detection of Lesions in CT Scans Using Intensity Distribution Supervision </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Shin%2C+S+Y">Seung Yeon Shin</a>, <a href="/search/cs?searchtype=author&query=Shen%2C+T+C">Thomas C. Shen</a>, <a href="/search/cs?searchtype=author&query=Summers%2C+R+M">Ronald M. Summers</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2307.05804v1-abstract-short" style="display: inline;"> We propose a method to incorporate the intensity information of a target lesion on CT scans in training segmentation and detection networks. We first build an intensity-based lesion probability (ILP) function from an intensity histogram of the target lesion. It is used to compute the probability of being the lesion for each voxel based on its intensity. Finally, the computed ILP map of each input… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2307.05804v1-abstract-full').style.display = 'inline'; document.getElementById('2307.05804v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2307.05804v1-abstract-full" style="display: none;"> We propose a method to incorporate the intensity information of a target lesion on CT scans in training segmentation and detection networks. We first build an intensity-based lesion probability (ILP) function from an intensity histogram of the target lesion. It is used to compute the probability of being the lesion for each voxel based on its intensity. Finally, the computed ILP map of each input CT scan is provided as additional supervision for network training, which aims to inform the network about possible lesion locations in terms of intensity values at no additional labeling cost. The method was applied to improve the segmentation of three different lesion types, namely, small bowel carcinoid tumor, kidney tumor, and lung nodule. The effectiveness of the proposed method on a detection task was also investigated. We observed improvements of 41.3% -> 47.8%, 74.2% -> 76.0%, and 26.4% -> 32.7% in segmenting small bowel carcinoid tumor, kidney tumor, and lung nodule, respectively, in terms of per case Dice scores. An improvement of 64.6% -> 75.5% was achieved in detecting kidney tumors in terms of average precision. The results of different usages of the ILP map and the effect of varied amount of training data are also presented. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2307.05804v1-abstract-full').style.display = 'none'; document.getElementById('2307.05804v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 11 July, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Computerized Medical Imaging and Graphics 2023</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2305.18277">arXiv:2305.18277</a> <span> [<a href="https://arxiv.org/pdf/2305.18277">pdf</a>, <a href="https://arxiv.org/format/2305.18277">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> 3DTeethSeg'22: 3D Teeth Scan Segmentation and Labeling Challenge </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Ben-Hamadou%2C+A">Achraf Ben-Hamadou</a>, <a href="/search/cs?searchtype=author&query=Smaoui%2C+O">Oussama Smaoui</a>, <a href="/search/cs?searchtype=author&query=Rekik%2C+A">Ahmed Rekik</a>, <a href="/search/cs?searchtype=author&query=Pujades%2C+S">Sergi Pujades</a>, <a href="/search/cs?searchtype=author&query=Boyer%2C+E">Edmond Boyer</a>, <a href="/search/cs?searchtype=author&query=Lim%2C+H">Hoyeon Lim</a>, <a href="/search/cs?searchtype=author&query=Kim%2C+M">Minchang Kim</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+M">Minkyung Lee</a>, <a href="/search/cs?searchtype=author&query=Chung%2C+M">Minyoung Chung</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yeong-Gil Shin</a>, <a href="/search/cs?searchtype=author&query=Leclercq%2C+M">Mathieu Leclercq</a>, <a href="/search/cs?searchtype=author&query=Cevidanes%2C+L">Lucia Cevidanes</a>, <a href="/search/cs?searchtype=author&query=Prieto%2C+J+C">Juan Carlos Prieto</a>, <a href="/search/cs?searchtype=author&query=Zhuang%2C+S">Shaojie Zhuang</a>, <a href="/search/cs?searchtype=author&query=Wei%2C+G">Guangshun Wei</a>, <a href="/search/cs?searchtype=author&query=Cui%2C+Z">Zhiming Cui</a>, <a href="/search/cs?searchtype=author&query=Zhou%2C+Y">Yuanfeng Zhou</a>, <a href="/search/cs?searchtype=author&query=Dascalu%2C+T">Tudor Dascalu</a>, <a href="/search/cs?searchtype=author&query=Ibragimov%2C+B">Bulat Ibragimov</a>, <a href="/search/cs?searchtype=author&query=Yong%2C+T">Tae-Hoon Yong</a>, <a href="/search/cs?searchtype=author&query=Ahn%2C+H">Hong-Gi Ahn</a>, <a href="/search/cs?searchtype=author&query=Kim%2C+W">Wan Kim</a>, <a href="/search/cs?searchtype=author&query=Han%2C+J">Jae-Hwan Han</a>, <a href="/search/cs?searchtype=author&query=Choi%2C+B">Byungsun Choi</a>, <a href="/search/cs?searchtype=author&query=van+Nistelrooij%2C+N">Niels van Nistelrooij</a> , et al. (7 additional authors not shown) </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2305.18277v1-abstract-short" style="display: inline;"> Teeth localization, segmentation, and labeling from intra-oral 3D scans are essential tasks in modern dentistry to enhance dental diagnostics, treatment planning, and population-based studies on oral health. However, developing automated algorithms for teeth analysis presents significant challenges due to variations in dental anatomy, imaging protocols, and limited availability of publicly accessi… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2305.18277v1-abstract-full').style.display = 'inline'; document.getElementById('2305.18277v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2305.18277v1-abstract-full" style="display: none;"> Teeth localization, segmentation, and labeling from intra-oral 3D scans are essential tasks in modern dentistry to enhance dental diagnostics, treatment planning, and population-based studies on oral health. However, developing automated algorithms for teeth analysis presents significant challenges due to variations in dental anatomy, imaging protocols, and limited availability of publicly accessible data. To address these challenges, the 3DTeethSeg'22 challenge was organized in conjunction with the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) in 2022, with a call for algorithms tackling teeth localization, segmentation, and labeling from intraoral 3D scans. A dataset comprising a total of 1800 scans from 900 patients was prepared, and each tooth was individually annotated by a human-machine hybrid algorithm. A total of 6 algorithms were evaluated on this dataset. In this study, we present the evaluation results of the 3DTeethSeg'22 challenge. The 3DTeethSeg'22 challenge code can be accessed at: https://github.com/abenhamadou/3DTeethSeg22_challenge <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2305.18277v1-abstract-full').style.display = 'none'; document.getElementById('2305.18277v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 29 May, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">29 pages, MICCAI 2022 Singapore, Satellite Event, Challenge</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2305.07097">arXiv:2305.07097</a> <span> [<a href="https://arxiv.org/pdf/2305.07097">pdf</a>, <a href="https://arxiv.org/format/2305.07097">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Software Engineering">cs.SE</span> </div> </div> <p class="title is-5 mathjax"> Automated Smell Detection and Recommendation in Natural Language Requirements </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Veizaga%2C+A">Alvaro Veizaga</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+S+Y">Seung Yeob Shin</a>, <a href="/search/cs?searchtype=author&query=Briand%2C+L+C">Lionel C. Briand</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2305.07097v2-abstract-short" style="display: inline;"> Requirement specifications are typically written in natural language (NL) due to its usability across multiple domains and understandability by all stakeholders. However, unstructured NL is prone to quality problems (e.g., ambiguity) when writing requirements, which can result in project failures. To address this issue, we present a tool, named Paska, that takes as input any NL requirements, autom… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2305.07097v2-abstract-full').style.display = 'inline'; document.getElementById('2305.07097v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2305.07097v2-abstract-full" style="display: none;"> Requirement specifications are typically written in natural language (NL) due to its usability across multiple domains and understandability by all stakeholders. However, unstructured NL is prone to quality problems (e.g., ambiguity) when writing requirements, which can result in project failures. To address this issue, we present a tool, named Paska, that takes as input any NL requirements, automatically detects quality problems as smells in the requirements, and offers recommendations to improve their quality. Our approach relies on natural language processing (NLP) techniques and a state-of-the-art controlled natural language (CNL) for requirements (Rimay), to detect smells and suggest recommendations using patterns defined in Rimay to improve requirement quality. We evaluated Paska through an industrial case study in the financial domain involving 13 systems and 2725 annotated requirements. The results show that our tool is accurate in detecting smells (89% precision and recall) and suggesting appropriate Rimay pattern recommendations (96% precision and 94% recall). <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2305.07097v2-abstract-full').style.display = 'none'; document.getElementById('2305.07097v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 25 November, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 11 May, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2304.01576">arXiv:2304.01576</a> <span> [<a href="https://arxiv.org/pdf/2304.01576">pdf</a>, <a href="https://arxiv.org/format/2304.01576">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Image and Video Processing">eess.IV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> </div> </div> <p class="title is-5 mathjax"> MESAHA-Net: Multi-Encoders based Self-Adaptive Hard Attention Network with Maximum Intensity Projections for Lung Nodule Segmentation in CT Scan </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Usman%2C+M">Muhammad Usman</a>, <a href="/search/cs?searchtype=author&query=Rehman%2C+A">Azka Rehman</a>, <a href="/search/cs?searchtype=author&query=Shahid%2C+A">Abdullah Shahid</a>, <a href="/search/cs?searchtype=author&query=Latif%2C+S">Siddique Latif</a>, <a href="/search/cs?searchtype=author&query=Byon%2C+S+S">Shi Sub Byon</a>, <a href="/search/cs?searchtype=author&query=Kim%2C+S+H">Sung Hyun Kim</a>, <a href="/search/cs?searchtype=author&query=Khan%2C+T+M">Tariq Mahmood Khan</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y+G">Yeong Gil Shin</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2304.01576v1-abstract-short" style="display: inline;"> Accurate lung nodule segmentation is crucial for early-stage lung cancer diagnosis, as it can substantially enhance patient survival rates. Computed tomography (CT) images are widely employed for early diagnosis in lung nodule analysis. However, the heterogeneity of lung nodules, size diversity, and the complexity of the surrounding environment pose challenges for developing robust nodule segmenta… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2304.01576v1-abstract-full').style.display = 'inline'; document.getElementById('2304.01576v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2304.01576v1-abstract-full" style="display: none;"> Accurate lung nodule segmentation is crucial for early-stage lung cancer diagnosis, as it can substantially enhance patient survival rates. Computed tomography (CT) images are widely employed for early diagnosis in lung nodule analysis. However, the heterogeneity of lung nodules, size diversity, and the complexity of the surrounding environment pose challenges for developing robust nodule segmentation methods. In this study, we propose an efficient end-to-end framework, the multi-encoder-based self-adaptive hard attention network (MESAHA-Net), for precise lung nodule segmentation in CT scans. MESAHA-Net comprises three encoding paths, an attention block, and a decoder block, facilitating the integration of three types of inputs: CT slice patches, forward and backward maximum intensity projection (MIP) images, and region of interest (ROI) masks encompassing the nodule. By employing a novel adaptive hard attention mechanism, MESAHA-Net iteratively performs slice-by-slice 2D segmentation of lung nodules, focusing on the nodule region in each slice to generate 3D volumetric segmentation of lung nodules. The proposed framework has been comprehensively evaluated on the LIDC-IDRI dataset, the largest publicly available dataset for lung nodule segmentation. The results demonstrate that our approach is highly robust for various lung nodule types, outperforming previous state-of-the-art techniques in terms of segmentation accuracy and computational complexity, rendering it suitable for real-time clinical implementation. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2304.01576v1-abstract-full').style.display = 'none'; document.getElementById('2304.01576v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 4 April, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2303.08329">arXiv:2303.08329</a> <span> [<a href="https://arxiv.org/pdf/2303.08329">pdf</a>, <a href="https://arxiv.org/format/2303.08329">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Sound">cs.SD</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computation and Language">cs.CL</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Audio and Speech Processing">eess.AS</span> </div> </div> <p class="title is-5 mathjax"> Cross-speaker Emotion Transfer by Manipulating Speech Style Latents </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Jo%2C+S">Suhee Jo</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+Y">Younggun Lee</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Yookyung Shin</a>, <a href="/search/cs?searchtype=author&query=Hwang%2C+Y">Yeongtae Hwang</a>, <a href="/search/cs?searchtype=author&query=Kim%2C+T">Taesu Kim</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2303.08329v1-abstract-short" style="display: inline;"> In recent years, emotional text-to-speech has shown considerable progress. However, it requires a large amount of labeled data, which is not easily accessible. Even if it is possible to acquire an emotional speech dataset, there is still a limitation in controlling emotion intensity. In this work, we propose a novel method for cross-speaker emotion transfer and manipulation using vector arithmetic… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2303.08329v1-abstract-full').style.display = 'inline'; document.getElementById('2303.08329v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2303.08329v1-abstract-full" style="display: none;"> In recent years, emotional text-to-speech has shown considerable progress. However, it requires a large amount of labeled data, which is not easily accessible. Even if it is possible to acquire an emotional speech dataset, there is still a limitation in controlling emotion intensity. In this work, we propose a novel method for cross-speaker emotion transfer and manipulation using vector arithmetic in latent style space. By leveraging only a few labeled samples, we generate emotional speech from reading-style speech without losing the speaker identity. Furthermore, emotion strength is readily controllable using a scalar value, providing an intuitive way for users to manipulate speech. Experimental results show the proposed method affords superior performance in terms of expressiveness, naturalness, and controllability, preserving speaker identity. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2303.08329v1-abstract-full').style.display = 'none'; document.getElementById('2303.08329v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 14 March, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">accepted to ICASSP 2023</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2303.05871">arXiv:2303.05871</a> <span> [<a href="https://arxiv.org/pdf/2303.05871">pdf</a>, <a href="https://arxiv.org/format/2303.05871">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> </div> <div class="is-inline-block" style="margin-left: 0.5rem"> <div class="tags has-addons"> <span class="tag is-dark is-size-7">doi</span> <span class="tag is-light is-size-7"><a class="" href="https://doi.org/10.1109/BIBM55620.2022.9995323">10.1109/BIBM55620.2022.9995323 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Accurate Real-time Polyp Detection in Videos from Concatenation of Latent Features Extracted from Consecutive Frames </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Qadir%2C+H+A">Hemin Ali Qadir</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Younghak Shin</a>, <a href="/search/cs?searchtype=author&query=Bergsland%2C+J">Jacob Bergsland</a>, <a href="/search/cs?searchtype=author&query=Balasingham%2C+I">Ilangko Balasingham</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2303.05871v1-abstract-short" style="display: inline;"> An efficient deep learning model that can be implemented in real-time for polyp detection is crucial to reducing polyp miss-rate during screening procedures. Convolutional neural networks (CNNs) are vulnerable to small changes in the input image. A CNN-based model may miss the same polyp appearing in a series of consecutive frames and produce unsubtle detection output due to changes in camera pose… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2303.05871v1-abstract-full').style.display = 'inline'; document.getElementById('2303.05871v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2303.05871v1-abstract-full" style="display: none;"> An efficient deep learning model that can be implemented in real-time for polyp detection is crucial to reducing polyp miss-rate during screening procedures. Convolutional neural networks (CNNs) are vulnerable to small changes in the input image. A CNN-based model may miss the same polyp appearing in a series of consecutive frames and produce unsubtle detection output due to changes in camera pose, lighting condition, light reflection, etc. In this study, we attempt to tackle this problem by integrating temporal information among neighboring frames. We propose an efficient feature concatenation method for a CNN-based encoder-decoder model without adding complexity to the model. The proposed method incorporates extracted feature maps of previous frames to detect polyps in the current frame. The experimental results demonstrate that the proposed method of feature concatenation improves the overall performance of automatic polyp detection in videos. The following results are obtained on a public video dataset: sensitivity 90.94\%, precision 90.53\%, and specificity 92.46% <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2303.05871v1-abstract-full').style.display = 'none'; document.getElementById('2303.05871v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 10 March, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> 2022 IEEE International Conference on Bioinformatics and Biomedicine (BIBM) (pp. 2461-2466). IEEE </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2302.13913">arXiv:2302.13913</a> <span> [<a href="https://arxiv.org/pdf/2302.13913">pdf</a>, <a href="https://arxiv.org/format/2302.13913">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Software Engineering">cs.SE</span> </div> <div class="is-inline-block" style="margin-left: 0.5rem"> <div class="tags has-addons"> <span class="tag is-dark is-size-7">doi</span> <span class="tag is-light is-size-7"><a class="" href="https://doi.org/10.1145/3624742">10.1145/3624742 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Stress Testing Control Loops in Cyber-Physical Systems </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Mandrioli%2C+C">Claudio Mandrioli</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+S+Y">Seung Yeob Shin</a>, <a href="/search/cs?searchtype=author&query=Maggio%2C+M">Martina Maggio</a>, <a href="/search/cs?searchtype=author&query=Bianculli%2C+D">Domenico Bianculli</a>, <a href="/search/cs?searchtype=author&query=Briand%2C+L">Lionel Briand</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2302.13913v4-abstract-short" style="display: inline;"> Cyber-Physical Systems (CPSs) are often safety-critical and deployed in uncertain environments. Identifying scenarios where CPSs do not comply with requirements is fundamental but difficult due to the multidisciplinary nature of CPSs. We investigate the testing of control-based CPSs, where control and software engineers develop the software collaboratively. Control engineers make design assumption… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2302.13913v4-abstract-full').style.display = 'inline'; document.getElementById('2302.13913v4-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2302.13913v4-abstract-full" style="display: none;"> Cyber-Physical Systems (CPSs) are often safety-critical and deployed in uncertain environments. Identifying scenarios where CPSs do not comply with requirements is fundamental but difficult due to the multidisciplinary nature of CPSs. We investigate the testing of control-based CPSs, where control and software engineers develop the software collaboratively. Control engineers make design assumptions during system development to leverage control theory and obtain guarantees on CPS behaviour. In the implemented system, however, such assumptions are not always satisfied, and their falsification can lead to loss of guarantees. We define stress testing of control-based CPSs as generating tests to falsify such design assumptions. We highlight different types of assumptions, focusing on the use of linearised physics models. To generate stress tests falsifying such assumptions, we leverage control theory to qualitatively characterise the input space of a control-based CPS. We propose a novel test parametrisation for control-based CPSs and use it with the input space characterisation to develop a stress testing approach. We evaluate our approach on three case study systems, including a drone, a continuous-current motor (in five configurations), and an aircraft.Our results show the effectiveness of the proposed testing approach in falsifying the design assumptions and highlighting the causes of assumption violations. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2302.13913v4-abstract-full').style.display = 'none'; document.getElementById('2302.13913v4-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 18 September, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 27 February, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Accepted for publication in August 2023 on the ACM Transactions on Software Engineering and Methodology (TOSEM)</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> ACM Trans. Softw. Eng. Methodol. 33, 2, Article 35 (February 2024), 58 pages </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2302.10288">arXiv:2302.10288</a> <span> [<a href="https://arxiv.org/pdf/2302.10288">pdf</a>, <a href="https://arxiv.org/format/2302.10288">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Software Engineering">cs.SE</span> </div> </div> <p class="title is-5 mathjax"> Probabilistic Safe WCET Estimation for Weakly Hard Real-Time Systems at Design Stages </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lee%2C+J">Jaekwon Lee</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+S+Y">Seung Yeob Shin</a>, <a href="/search/cs?searchtype=author&query=Briand%2C+L">Lionel Briand</a>, <a href="/search/cs?searchtype=author&query=Nejati%2C+S">Shiva Nejati</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2302.10288v3-abstract-short" style="display: inline;"> Weakly hard real-time systems can, to some degree, tolerate deadline misses, but their schedulability still needs to be analyzed to ensure their quality of service. Such analysis usually occurs at early design stages to provide implementation guidelines to engineers so that they can make better design decisions. Estimating worst-case execution times (WCET) is a key input to schedulability analysis… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2302.10288v3-abstract-full').style.display = 'inline'; document.getElementById('2302.10288v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2302.10288v3-abstract-full" style="display: none;"> Weakly hard real-time systems can, to some degree, tolerate deadline misses, but their schedulability still needs to be analyzed to ensure their quality of service. Such analysis usually occurs at early design stages to provide implementation guidelines to engineers so that they can make better design decisions. Estimating worst-case execution times (WCET) is a key input to schedulability analysis. However, early on during system design, estimating WCET values is challenging and engineers usually determine them as plausible ranges based on their domain knowledge. Our approach aims at finding restricted, safe WCET sub-ranges given a set of ranges initially estimated by experts in the context of weakly hard real-time systems. To this end, we leverage (1) multi-objective search aiming at maximizing the violation of weakly hard constraints in order to find worst-case scheduling scenarios and (2) polynomial logistic regression to infer safe WCET ranges with a probabilistic interpretation. We evaluated our approach by applying it to an industrial system in the satellite domain and several realistic synthetic systems. The results indicate that our approach significantly outperforms a baseline relying on random search without learning, and estimates safe WCET ranges with a high degree of confidence in practical time (< 23h). <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2302.10288v3-abstract-full').style.display = 'none'; document.getElementById('2302.10288v3-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 11 August, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 20 February, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Accepted for publication by TOSEM (in Aug 2023)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2302.09835">arXiv:2302.09835</a> <span> [<a href="https://arxiv.org/pdf/2302.09835">pdf</a>, <a href="https://arxiv.org/ps/2302.09835">ps</a>, <a href="https://arxiv.org/format/2302.09835">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Image and Video Processing">eess.IV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> <div class="is-inline-block" style="margin-left: 0.5rem"> <div class="tags has-addons"> <span class="tag is-dark is-size-7">doi</span> <span class="tag is-light is-size-7"><a class="" href="https://doi.org/10.1016/j.bspc.2022.103491">10.1016/j.bspc.2022.103491 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Simple U-net Based Synthetic Polyp Image Generation: Polyp to Negative and Negative to Polyp </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Qadir%2C+H+A">Hemin Ali Qadir</a>, <a href="/search/cs?searchtype=author&query=Balasingham%2C+I">Ilangko Balasingham</a>, <a href="/search/cs?searchtype=author&query=Shin%2C+Y">Younghak Shin</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2302.09835v1-abstract-short" style="display: inline;"> Synthetic polyp generation is a good alternative to overcome the privacy problem of medical data and the lack of various polyp samples. In this study, we propose a deep learning-based polyp image generation framework that generates synthetic polyp images that are similar to real ones. We suggest a framework that converts a given polyp image into a negative image (image without a polyp) using a sim… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2302.09835v1-abstract-full').style.display = 'inline'; document.getElementById('2302.09835v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2302.09835v1-abstract-full" style="display: none;"> Synthetic polyp generation is a good alternative to overcome the privacy problem of medical data and the lack of various polyp samples. In this study, we propose a deep learning-based polyp image generation framework that generates synthetic polyp images that are similar to real ones. We suggest a framework that converts a given polyp image into a negative image (image without a polyp) using a simple conditional GAN architecture and then converts the negative image into a new-looking polyp image using the same network. In addition, by using the controllable polyp masks, polyps with various characteristics can be generated from one input condition. The generated polyp images can be used directly as training images for polyp detection and segmentation without additional labeling. To quantitatively assess the quality of generated synthetic polyps, we use public polyp image and video datasets combined with the generated synthetic images to examine the performance improvement of several detection and segmentation models. Experimental results show that we obtain performance gains when the generated polyp images are added to the training set. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2302.09835v1-abstract-full').style.display = 'none'; document.getElementById('2302.09835v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 20 February, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2023. </p> </li> </ol> <nav class="pagination is-small is-centered breathe-horizontal" role="navigation" aria-label="pagination"> <a href="" class="pagination-previous is-invisible">Previous </a> <a href="/search/?searchtype=author&query=Shin%2C+Y&start=50" class="pagination-next" >Next </a> <ul class="pagination-list"> <li> <a href="/search/?searchtype=author&query=Shin%2C+Y&start=0" class="pagination-link is-current" aria-label="Goto page 1">1 </a> </li> <li> <a href="/search/?searchtype=author&query=Shin%2C+Y&start=50" class="pagination-link " aria-label="Page 2" aria-current="page">2 </a> </li> <li> <a href="/search/?searchtype=author&query=Shin%2C+Y&start=100" class="pagination-link " aria-label="Page 3" aria-current="page">3 </a> </li> <li> <a href="/search/?searchtype=author&query=Shin%2C+Y&start=150" class="pagination-link " aria-label="Page 4" aria-current="page">4 </a> </li> </ul> </nav> <div class="is-hidden-tablet">  <span class="help" style="display: inline-block;"><a href="https://github.com/arXiv/arxiv-search/releases">Search v0.5.6 released 2020-02-24</a>  </span> </div> </div> </main> <footer> <div class="columns is-desktop" role="navigation" aria-label="Secondary">  <div class="column"> <div class="columns"> <div class="column"> <ul class="nav-spaced"> <li><a href="https://info.arxiv.org/about">About</a></li> <li><a href="https://info.arxiv.org/help">Help</a></li> </ul> </div> <div class="column"> <ul class="nav-spaced"> <li> <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512" class="icon filter-black" role="presentation"><title>contact arXiv</title><desc>Click here to contact arXiv</desc><path d="M502.3 190.8c3.9-3.1 9.7-.2 9.7 4.7V400c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V195.6c0-5 5.7-7.8 9.7-4.7 22.4 17.4 52.1 39.5 154.1 113.6 21.1 15.4 56.7 47.8 92.2 47.6 35.7.3 72-32.8 92.3-47.6 102-74.1 131.6-96.3 154-113.7zM256 320c23.2.4 56.6-29.2 73.4-41.4 132.7-96.3 142.8-104.7 173.4-128.7 5.8-4.5 9.2-11.5 9.2-18.9v-19c0-26.5-21.5-48-48-48H48C21.5 64 0 85.5 0 112v19c0 7.4 3.4 14.3 9.2 18.9 30.6 23.9 40.7 32.4 173.4 128.7 16.8 12.2 50.2 41.8 73.4 41.4z"/></svg> <a href="https://info.arxiv.org/help/contact.html"> Contact</a> </li> <li> <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512" class="icon filter-black" role="presentation"><title>subscribe to arXiv mailings</title><desc>Click here to subscribe</desc><path d="M476 3.2L12.5 270.6c-18.1 10.4-15.8 35.6 2.2 43.2L121 358.4l287.3-253.2c5.5-4.9 13.3 2.6 8.6 8.3L176 407v80.5c0 23.6 28.5 32.9 42.5 15.8L282 426l124.6 52.2c14.2 6 30.4-2.9 33-18.2l72-432C515 7.8 493.3-6.8 476 3.2z"/></svg> <a href="https://info.arxiv.org/help/subscribe"> Subscribe</a> </li> </ul> </div> </div> </div>   <div class="column"> <div class="columns"> <div class="column"> <ul class="nav-spaced"> <li><a href="https://info.arxiv.org/help/license/index.html">Copyright</a></li> <li><a href="https://info.arxiv.org/help/policies/privacy_policy.html">Privacy Policy</a></li> </ul> </div> <div class="column sorry-app-links"> <ul class="nav-spaced"> <li><a href="https://info.arxiv.org/help/web_accessibility.html">Web Accessibility Assistance</a></li> <li> <p class="help"> <a class="a11y-main-link" href="https://status.arxiv.org" target="_blank">arXiv Operational Status <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 256 512" class="icon filter-dark_grey" role="presentation"><path d="M224.3 273l-136 136c-9.4 9.4-24.6 9.4-33.9 0l-22.6-22.6c-9.4-9.4-9.4-24.6 0-33.9l96.4-96.4-96.4-96.4c-9.4-9.4-9.4-24.6 0-33.9L54.3 103c9.4-9.4 24.6-9.4 33.9 0l136 136c9.5 9.4 9.5 24.6.1 34z"/></svg></a><br> Get status notifications via <a class="is-link" href="https://subscribe.sorryapp.com/24846f03/email/new" target="_blank"><svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512" class="icon filter-black" role="presentation"><path d="M502.3 190.8c3.9-3.1 9.7-.2 9.7 4.7V400c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V195.6c0-5 5.7-7.8 9.7-4.7 22.4 17.4 52.1 39.5 154.1 113.6 21.1 15.4 56.7 47.8 92.2 47.6 35.7.3 72-32.8 92.3-47.6 102-74.1 131.6-96.3 154-113.7zM256 320c23.2.4 56.6-29.2 73.4-41.4 132.7-96.3 142.8-104.7 173.4-128.7 5.8-4.5 9.2-11.5 9.2-18.9v-19c0-26.5-21.5-48-48-48H48C21.5 64 0 85.5 0 112v19c0 7.4 3.4 14.3 9.2 18.9 30.6 23.9 40.7 32.4 173.4 128.7 16.8 12.2 50.2 41.8 73.4 41.4z"/></svg>email</a> or <a class="is-link" href="https://subscribe.sorryapp.com/24846f03/slack/new" target="_blank"><svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 448 512" class="icon filter-black" role="presentation"><path d="M94.12 315.1c0 25.9-21.16 47.06-47.06 47.06S0 341 0 315.1c0-25.9 21.16-47.06 47.06-47.06h47.06v47.06zm23.72 0c0-25.9 21.16-47.06 47.06-47.06s47.06 21.16 47.06 47.06v117.84c0 25.9-21.16 47.06-47.06 47.06s-47.06-21.16-47.06-47.06V315.1zm47.06-188.98c-25.9 0-47.06-21.16-47.06-47.06S139 32 164.9 32s47.06 21.16 47.06 47.06v47.06H164.9zm0 23.72c25.9 0 47.06 21.16 47.06 47.06s-21.16 47.06-47.06 47.06H47.06C21.16 243.96 0 222.8 0 196.9s21.16-47.06 47.06-47.06H164.9zm188.98 47.06c0-25.9 21.16-47.06 47.06-47.06 25.9 0 47.06 21.16 47.06 47.06s-21.16 47.06-47.06 47.06h-47.06V196.9zm-23.72 0c0 25.9-21.16 47.06-47.06 47.06-25.9 0-47.06-21.16-47.06-47.06V79.06c0-25.9 21.16-47.06 47.06-47.06 25.9 0 47.06 21.16 47.06 47.06V196.9zM283.1 385.88c25.9 0 47.06 21.16 47.06 47.06 0 25.9-21.16 47.06-47.06 47.06-25.9 0-47.06-21.16-47.06-47.06v-47.06h47.06zm0-23.72c-25.9 0-47.06-21.16-47.06-47.06 0-25.9 21.16-47.06 47.06-47.06h117.84c25.9 0 47.06 21.16 47.06 47.06 0 25.9-21.16 47.06-47.06 47.06H283.1z"/></svg>slack</a> </p> </li> </ul> </div> </div> </div>  </div> </footer> <script src="https://static.arxiv.org/static/base/1.0.0a5/js/member_acknowledgement.js"></script> </body> </html>

CINXE.COM

Search | arXiv e-print repository