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 1–40 of 40 results for author: <span class="mathjax">Law, M</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=Law%2C+M">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="Law, M"> </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=Law%2C+M&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="Law, M"> <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> <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/2411.12915">arXiv:2411.12915</a> <span> [<a href="https://arxiv.org/pdf/2411.12915">pdf</a>, <a href="https://arxiv.org/format/2411.12915">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"> VILA-M3: Enhancing Vision-Language Models with Medical Expert Knowledge </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Nath%2C+V">Vishwesh Nath</a>, <a href="/search/cs?searchtype=author&query=Li%2C+W">Wenqi Li</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+D">Dong Yang</a>, <a href="/search/cs?searchtype=author&query=Myronenko%2C+A">Andriy Myronenko</a>, <a href="/search/cs?searchtype=author&query=Zheng%2C+M">Mingxin Zheng</a>, <a href="/search/cs?searchtype=author&query=Lu%2C+Y">Yao Lu</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+Z">Zhijian Liu</a>, <a href="/search/cs?searchtype=author&query=Yin%2C+H">Hongxu Yin</a>, <a href="/search/cs?searchtype=author&query=Law%2C+Y+M">Yee Man Law</a>, <a href="/search/cs?searchtype=author&query=Tang%2C+Y">Yucheng Tang</a>, <a href="/search/cs?searchtype=author&query=Guo%2C+P">Pengfei Guo</a>, <a href="/search/cs?searchtype=author&query=Zhao%2C+C">Can Zhao</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+Z">Ziyue Xu</a>, <a href="/search/cs?searchtype=author&query=He%2C+Y">Yufan He</a>, <a href="/search/cs?searchtype=author&query=Heinrich%2C+G">Greg Heinrich</a>, <a href="/search/cs?searchtype=author&query=Aylward%2C+S">Stephen Aylward</a>, <a href="/search/cs?searchtype=author&query=Edgar%2C+M">Marc Edgar</a>, <a href="/search/cs?searchtype=author&query=Zephyr%2C+M">Michael Zephyr</a>, <a href="/search/cs?searchtype=author&query=Molchanov%2C+P">Pavlo Molchanov</a>, <a href="/search/cs?searchtype=author&query=Turkbey%2C+B">Baris Turkbey</a>, <a href="/search/cs?searchtype=author&query=Roth%2C+H">Holger Roth</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+D">Daguang Xu</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="2411.12915v1-abstract-short" style="display: inline;"> Generalist vision language models (VLMs) have made significant strides in computer vision, but they fall short in specialized fields like healthcare, where expert knowledge is essential. In traditional computer vision tasks, creative or approximate answers may be acceptable, but in healthcare, precision is paramount.Current large multimodal models like Gemini and GPT-4o are insufficient for medica… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.12915v1-abstract-full').style.display = 'inline'; document.getElementById('2411.12915v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.12915v1-abstract-full" style="display: none;"> Generalist vision language models (VLMs) have made significant strides in computer vision, but they fall short in specialized fields like healthcare, where expert knowledge is essential. In traditional computer vision tasks, creative or approximate answers may be acceptable, but in healthcare, precision is paramount.Current large multimodal models like Gemini and GPT-4o are insufficient for medical tasks due to their reliance on memorized internet knowledge rather than the nuanced expertise required in healthcare. VLMs are usually trained in three stages: vision pre-training, vision-language pre-training, and instruction fine-tuning (IFT). IFT has been typically applied using a mixture of generic and healthcare data. In contrast, we propose that for medical VLMs, a fourth stage of specialized IFT is necessary, which focuses on medical data and includes information from domain expert models. Domain expert models developed for medical use are crucial because they are specifically trained for certain clinical tasks, e.g. to detect tumors and classify abnormalities through segmentation and classification, which learn fine-grained features of medical data$-$features that are often too intricate for a VLM to capture effectively especially in radiology. This paper introduces a new framework, VILA-M3, for medical VLMs that utilizes domain knowledge via expert models. Through our experiments, we show an improved state-of-the-art (SOTA) performance with an average improvement of ~9% over the prior SOTA model Med-Gemini and ~6% over models trained on the specific tasks. Our approach emphasizes the importance of domain expertise in creating precise, reliable VLMs for medical applications. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.12915v1-abstract-full').style.display = 'none'; document.getElementById('2411.12915v1-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> 19 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/2410.23910">arXiv:2410.23910</a> <span> [<a href="https://arxiv.org/pdf/2410.23910">pdf</a>, <a href="https://arxiv.org/format/2410.23910">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"> Uncertainty Estimation for 3D Object Detection via Evidential Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Durasov%2C+N">Nikita Durasov</a>, <a href="/search/cs?searchtype=author&query=Mahmood%2C+R">Rafid Mahmood</a>, <a href="/search/cs?searchtype=author&query=Choi%2C+J">Jiwoong Choi</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M+T">Marc T. Law</a>, <a href="/search/cs?searchtype=author&query=Lucas%2C+J">James Lucas</a>, <a href="/search/cs?searchtype=author&query=Fua%2C+P">Pascal Fua</a>, <a href="/search/cs?searchtype=author&query=Alvarez%2C+J+M">Jose M. Alvarez</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="2410.23910v1-abstract-short" style="display: inline;"> 3D object detection is an essential task for computer vision applications in autonomous vehicles and robotics. However, models often struggle to quantify detection reliability, leading to poor performance on unfamiliar scenes. We introduce a framework for quantifying uncertainty in 3D object detection by leveraging an evidential learning loss on Bird's Eye View representations in the 3D detector.… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.23910v1-abstract-full').style.display = 'inline'; document.getElementById('2410.23910v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2410.23910v1-abstract-full" style="display: none;"> 3D object detection is an essential task for computer vision applications in autonomous vehicles and robotics. However, models often struggle to quantify detection reliability, leading to poor performance on unfamiliar scenes. We introduce a framework for quantifying uncertainty in 3D object detection by leveraging an evidential learning loss on Bird's Eye View representations in the 3D detector. These uncertainty estimates require minimal computational overhead and are generalizable across different architectures. We demonstrate both the efficacy and importance of these uncertainty estimates on identifying out-of-distribution scenes, poorly localized objects, and missing (false negative) detections; our framework consistently improves over baselines by 10-20% on average. Finally, we integrate this suite of tasks into a system where a 3D object detector auto-labels driving scenes and our uncertainty estimates verify label correctness before the labels are used to train a second model. Here, our uncertainty-driven verification results in a 1% improvement in mAP and a 1-2% improvement in NDS. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.23910v1-abstract-full').style.display = 'none'; document.getElementById('2410.23910v1-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, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2409.20562">arXiv:2409.20562</a> <span> [<a href="https://arxiv.org/pdf/2409.20562">pdf</a>, <a href="https://arxiv.org/format/2409.20562">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="Graphics">cs.GR</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/10.1145/3680528.3687634">10.1145/3680528.3687634 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> SpaceMesh: A Continuous Representation for Learning Manifold Surface Meshes </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Shen%2C+T">Tianchang Shen</a>, <a href="/search/cs?searchtype=author&query=Li%2C+Z">Zhaoshuo Li</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M">Marc Law</a>, <a href="/search/cs?searchtype=author&query=Atzmon%2C+M">Matan Atzmon</a>, <a href="/search/cs?searchtype=author&query=Fidler%2C+S">Sanja Fidler</a>, <a href="/search/cs?searchtype=author&query=Lucas%2C+J">James Lucas</a>, <a href="/search/cs?searchtype=author&query=Gao%2C+J">Jun Gao</a>, <a href="/search/cs?searchtype=author&query=Sharp%2C+N">Nicholas Sharp</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="2409.20562v1-abstract-short" style="display: inline;"> Meshes are ubiquitous in visual computing and simulation, yet most existing machine learning techniques represent meshes only indirectly, e.g. as the level set of a scalar field or deformation of a template, or as a disordered triangle soup lacking local structure. This work presents a scheme to directly generate manifold, polygonal meshes of complex connectivity as the output of a neural network.… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2409.20562v1-abstract-full').style.display = 'inline'; document.getElementById('2409.20562v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2409.20562v1-abstract-full" style="display: none;"> Meshes are ubiquitous in visual computing and simulation, yet most existing machine learning techniques represent meshes only indirectly, e.g. as the level set of a scalar field or deformation of a template, or as a disordered triangle soup lacking local structure. This work presents a scheme to directly generate manifold, polygonal meshes of complex connectivity as the output of a neural network. Our key innovation is to define a continuous latent connectivity space at each mesh vertex, which implies the discrete mesh. In particular, our vertex embeddings generate cyclic neighbor relationships in a halfedge mesh representation, which gives a guarantee of edge-manifoldness and the ability to represent general polygonal meshes. This representation is well-suited to machine learning and stochastic optimization, without restriction on connectivity or topology. We first explore the basic properties of this representation, then use it to fit distributions of meshes from large datasets. The resulting models generate diverse meshes with tessellation structure learned from the dataset population, with concise details and high-quality mesh elements. In applications, this approach not only yields high-quality outputs from generative models, but also enables directly learning challenging geometry processing tasks such as mesh repair. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2409.20562v1-abstract-full').style.display = 'none'; document.getElementById('2409.20562v1-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 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">published at SIGGRAPH Asia 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/2409.00585">arXiv:2409.00585</a> <span> [<a href="https://arxiv.org/pdf/2409.00585">pdf</a>, <a href="https://arxiv.org/format/2409.00585">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"> McCaD: Multi-Contrast MRI Conditioned, Adaptive Adversarial Diffusion Model for High-Fidelity MRI Synthesis </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Dayarathna%2C+S">Sanuwani Dayarathna</a>, <a href="/search/cs?searchtype=author&query=Islam%2C+K+T">Kh Tohidul Islam</a>, <a href="/search/cs?searchtype=author&query=Zhuang%2C+B">Bohan Zhuang</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+G">Guang Yang</a>, <a href="/search/cs?searchtype=author&query=Cai%2C+J">Jianfei Cai</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M">Meng Law</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+Z">Zhaolin Chen</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="2409.00585v1-abstract-short" style="display: inline;"> Magnetic Resonance Imaging (MRI) is instrumental in clinical diagnosis, offering diverse contrasts that provide comprehensive diagnostic information. However, acquiring multiple MRI contrasts is often constrained by high costs, long scanning durations, and patient discomfort. Current synthesis methods, typically focused on single-image contrasts, fall short in capturing the collective nuances acro… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2409.00585v1-abstract-full').style.display = 'inline'; document.getElementById('2409.00585v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2409.00585v1-abstract-full" style="display: none;"> Magnetic Resonance Imaging (MRI) is instrumental in clinical diagnosis, offering diverse contrasts that provide comprehensive diagnostic information. However, acquiring multiple MRI contrasts is often constrained by high costs, long scanning durations, and patient discomfort. Current synthesis methods, typically focused on single-image contrasts, fall short in capturing the collective nuances across various contrasts. Moreover, existing methods for multi-contrast MRI synthesis often fail to accurately map feature-level information across multiple imaging contrasts. We introduce McCaD (Multi-Contrast MRI Conditioned Adaptive Adversarial Diffusion), a novel framework leveraging an adversarial diffusion model conditioned on multiple contrasts for high-fidelity MRI synthesis. McCaD significantly enhances synthesis accuracy by employing a multi-scale, feature-guided mechanism, incorporating denoising and semantic encoders. An adaptive feature maximization strategy and a spatial feature-attentive loss have been introduced to capture more intrinsic features across multiple contrasts. This facilitates a precise and comprehensive feature-guided denoising process. Extensive experiments on tumor and healthy multi-contrast MRI datasets demonstrated that the McCaD outperforms state-of-the-art baselines quantitively and qualitatively. The code is provided with supplementary materials. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2409.00585v1-abstract-full').style.display = 'none'; document.getElementById('2409.00585v1-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 August, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2408.13885">arXiv:2408.13885</a> <span> [<a href="https://arxiv.org/pdf/2408.13885">pdf</a>, <a href="https://arxiv.org/format/2408.13885">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="Discrete Mathematics">cs.DM</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="Metric Geometry">math.MG</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"> Neural Spacetimes for DAG Representation Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Borde%2C+H+S+d+O">Haitz S谩ez de Oc谩riz Borde</a>, <a href="/search/cs?searchtype=author&query=Kratsios%2C+A">Anastasis Kratsios</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M+T">Marc T. Law</a>, <a href="/search/cs?searchtype=author&query=Dong%2C+X">Xiaowen Dong</a>, <a href="/search/cs?searchtype=author&query=Bronstein%2C+M">Michael Bronstein</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="2408.13885v1-abstract-short" style="display: inline;"> We propose a class of trainable deep learning-based geometries called Neural Spacetimes (NSTs), which can universally represent nodes in weighted directed acyclic graphs (DAGs) as events in a spacetime manifold. While most works in the literature focus on undirected graph representation learning or causality embedding separately, our differentiable geometry can encode both graph edge weights in it… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2408.13885v1-abstract-full').style.display = 'inline'; document.getElementById('2408.13885v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2408.13885v1-abstract-full" style="display: none;"> We propose a class of trainable deep learning-based geometries called Neural Spacetimes (NSTs), which can universally represent nodes in weighted directed acyclic graphs (DAGs) as events in a spacetime manifold. While most works in the literature focus on undirected graph representation learning or causality embedding separately, our differentiable geometry can encode both graph edge weights in its spatial dimensions and causality in the form of edge directionality in its temporal dimensions. We use a product manifold that combines a quasi-metric (for space) and a partial order (for time). NSTs are implemented as three neural networks trained in an end-to-end manner: an embedding network, which learns to optimize the location of nodes as events in the spacetime manifold, and two other networks that optimize the space and time geometries in parallel, which we call a neural (quasi-)metric and a neural partial order, respectively. The latter two networks leverage recent ideas at the intersection of fractal geometry and deep learning to shape the geometry of the representation space in a data-driven fashion, unlike other works in the literature that use fixed spacetime manifolds such as Minkowski space or De Sitter space to embed DAGs. Our main theoretical guarantee is a universal embedding theorem, showing that any $k$-point DAG can be embedded into an NST with $1+\mathcal{O}(\log(k))$ distortion while exactly preserving its causal structure. The total number of parameters defining the NST is sub-cubic in $k$ and linear in the width of the DAG. If the DAG has a planar Hasse diagram, this is improved to $\mathcal{O}(\log(k)) + 2)$ spatial and 2 temporal dimensions. We validate our framework computationally with synthetic weighted DAGs and real-world network embeddings; in both cases, the NSTs achieve lower embedding distortions than their counterparts using fixed spacetime geometries. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2408.13885v1-abstract-full').style.display = 'none'; document.getElementById('2408.13885v1-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 August, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> August 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">12 pages: main body and 19 pages: appendix</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2405.19595">arXiv:2405.19595</a> <span> [<a href="https://arxiv.org/pdf/2405.19595">pdf</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"> The RSNA Abdominal Traumatic Injury CT (RATIC) Dataset </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Rudie%2C+J+D">Jeffrey D. Rudie</a>, <a href="/search/cs?searchtype=author&query=Lin%2C+H">Hui-Ming Lin</a>, <a href="/search/cs?searchtype=author&query=Ball%2C+R+L">Robyn L. Ball</a>, <a href="/search/cs?searchtype=author&query=Jalal%2C+S">Sabeena Jalal</a>, <a href="/search/cs?searchtype=author&query=Prevedello%2C+L+M">Luciano M. Prevedello</a>, <a href="/search/cs?searchtype=author&query=Nicolaou%2C+S">Savvas Nicolaou</a>, <a href="/search/cs?searchtype=author&query=Marinelli%2C+B+S">Brett S. Marinelli</a>, <a href="/search/cs?searchtype=author&query=Flanders%2C+A+E">Adam E. Flanders</a>, <a href="/search/cs?searchtype=author&query=Magudia%2C+K">Kirti Magudia</a>, <a href="/search/cs?searchtype=author&query=Shih%2C+G">George Shih</a>, <a href="/search/cs?searchtype=author&query=Davis%2C+M+A">Melissa A. Davis</a>, <a href="/search/cs?searchtype=author&query=Mongan%2C+J">John Mongan</a>, <a href="/search/cs?searchtype=author&query=Chang%2C+P+D">Peter D. Chang</a>, <a href="/search/cs?searchtype=author&query=Berger%2C+F+H">Ferco H. Berger</a>, <a href="/search/cs?searchtype=author&query=Hermans%2C+S">Sebastiaan Hermans</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M">Meng Law</a>, <a href="/search/cs?searchtype=author&query=Richards%2C+T">Tyler Richards</a>, <a href="/search/cs?searchtype=author&query=Grunz%2C+J">Jan-Peter Grunz</a>, <a href="/search/cs?searchtype=author&query=Kunz%2C+A+S">Andreas Steven Kunz</a>, <a href="/search/cs?searchtype=author&query=Mathur%2C+S">Shobhit Mathur</a>, <a href="/search/cs?searchtype=author&query=Galea-Soler%2C+S">Sandro Galea-Soler</a>, <a href="/search/cs?searchtype=author&query=Chung%2C+A+D">Andrew D. Chung</a>, <a href="/search/cs?searchtype=author&query=Afat%2C+S">Saif Afat</a>, <a href="/search/cs?searchtype=author&query=Kuo%2C+C">Chin-Chi Kuo</a>, <a href="/search/cs?searchtype=author&query=Aweidah%2C+L">Layal Aweidah</a> , et al. (15 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="2405.19595v1-abstract-short" style="display: inline;"> The RSNA Abdominal Traumatic Injury CT (RATIC) dataset is the largest publicly available collection of adult abdominal CT studies annotated for traumatic injuries. This dataset includes 4,274 studies from 23 institutions across 14 countries. The dataset is freely available for non-commercial use via Kaggle at https://www.kaggle.com/competitions/rsna-2023-abdominal-trauma-detection. Created for the… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2405.19595v1-abstract-full').style.display = 'inline'; document.getElementById('2405.19595v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2405.19595v1-abstract-full" style="display: none;"> The RSNA Abdominal Traumatic Injury CT (RATIC) dataset is the largest publicly available collection of adult abdominal CT studies annotated for traumatic injuries. This dataset includes 4,274 studies from 23 institutions across 14 countries. The dataset is freely available for non-commercial use via Kaggle at https://www.kaggle.com/competitions/rsna-2023-abdominal-trauma-detection. Created for the RSNA 2023 Abdominal Trauma Detection competition, the dataset encourages the development of advanced machine learning models for detecting abdominal injuries on CT scans. The dataset encompasses detection and classification of traumatic injuries across multiple organs, including the liver, spleen, kidneys, bowel, and mesentery. Annotations were created by expert radiologists from the American Society of Emergency Radiology (ASER) and Society of Abdominal Radiology (SAR). The dataset is annotated at multiple levels, including the presence of injuries in three solid organs with injury grading, image-level annotations for active extravasations and bowel injury, and voxelwise segmentations of each of the potentially injured organs. With the release of this dataset, we hope to facilitate research and development in machine learning and abdominal trauma that can lead to improved patient care and outcomes. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2405.19595v1-abstract-full').style.display = 'none'; document.getElementById('2405.19595v1-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, 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">40 pages, 2 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/2405.08337">arXiv:2405.08337</a> <span> [<a href="https://arxiv.org/pdf/2405.08337">pdf</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"> Perivascular space Identification Nnunet for Generalised Usage (PINGU) </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Sinclair%2C+B">Benjamin Sinclair</a>, <a href="/search/cs?searchtype=author&query=Vivash%2C+L">Lucy Vivash</a>, <a href="/search/cs?searchtype=author&query=Moses%2C+J">Jasmine Moses</a>, <a href="/search/cs?searchtype=author&query=Lynch%2C+M">Miranda Lynch</a>, <a href="/search/cs?searchtype=author&query=Pham%2C+W">William Pham</a>, <a href="/search/cs?searchtype=author&query=Dorfman%2C+K">Karina Dorfman</a>, <a href="/search/cs?searchtype=author&query=Marotta%2C+C">Cassandra Marotta</a>, <a href="/search/cs?searchtype=author&query=Koh%2C+S">Shaun Koh</a>, <a href="/search/cs?searchtype=author&query=Bunyamin%2C+J">Jacob Bunyamin</a>, <a href="/search/cs?searchtype=author&query=Rowsthorn%2C+E">Ella Rowsthorn</a>, <a href="/search/cs?searchtype=author&query=Jarema%2C+A">Alex Jarema</a>, <a href="/search/cs?searchtype=author&query=Peiris%2C+H">Himashi Peiris</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+Z">Zhaolin Chen</a>, <a href="/search/cs?searchtype=author&query=Shultz%2C+S+R">Sandy R Shultz</a>, <a href="/search/cs?searchtype=author&query=Wright%2C+D+K">David K Wright</a>, <a href="/search/cs?searchtype=author&query=Kong%2C+D">Dexiao Kong</a>, <a href="/search/cs?searchtype=author&query=Naismith%2C+S+L">Sharon L. Naismith</a>, <a href="/search/cs?searchtype=author&query=OBrien%2C+T+J">Terence J. OBrien</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M">Meng Law</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="2405.08337v2-abstract-short" style="display: inline;"> Perivascular spaces(PVSs) form a central component of the brain艣 waste clearance system, the glymphatic system. These structures are visible on MRI images, and their morphology is associated with aging and neurological disease. Manual quantification of PVS is time consuming and subjective. Numerous deep learning methods for PVS segmentation have been developed, however the majority have been devel… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2405.08337v2-abstract-full').style.display = 'inline'; document.getElementById('2405.08337v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2405.08337v2-abstract-full" style="display: none;"> Perivascular spaces(PVSs) form a central component of the brain艣 waste clearance system, the glymphatic system. These structures are visible on MRI images, and their morphology is associated with aging and neurological disease. Manual quantification of PVS is time consuming and subjective. Numerous deep learning methods for PVS segmentation have been developed, however the majority have been developed and evaluated on homogenous datasets and high resolution scans, perhaps limiting their applicability for the wide range of image qualities acquired in clinic and research. In this work we train a nnUNet, a top-performing biomedical image segmentation algorithm, on a heterogenous training sample of manually segmented MRI images of a range of different qualities and resolutions from 6 different datasets. These are compared to publicly available deep learning methods for 3D segmentation of PVS. The resulting model, PINGU (Perivascular space Identification Nnunet for Generalised Usage), achieved voxel and cluster level dice scores of 0.50(SD=0.15), 0.63(0.17) in the white matter(WM), and 0.54(0.11), 0.66(0.17) in the basal ganglia(BG). Performance on data from unseen sites was substantially lower for both PINGU(0.20-0.38(WM, voxel), 0.29-0.58(WM, cluster), 0.22-0.36(BG, voxel), 0.46-0.60(BG, cluster)) and the publicly available algorithms(0.18-0.30(WM, voxel), 0.29-0.38(WM cluster), 0.10-0.20(BG, voxel), 0.15-0.37(BG, cluster)), but PINGU strongly outperformed the publicly available algorithms, particularly in the BG. Finally, training PINGU on manual segmentations from a single site with homogenous scan properties gave marginally lower performances on internal cross-validation, but in some cases gave higher performance on external validation. PINGU stands out as broad-use PVS segmentation tool, with particular strength in the BG, an area of PVS related to vascular disease and pathology. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2405.08337v2-abstract-full').style.display = 'none'; document.getElementById('2405.08337v2-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 May, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 14 May, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2402.03460">arXiv:2402.03460</a> <span> [<a href="https://arxiv.org/pdf/2402.03460">pdf</a>, <a href="https://arxiv.org/format/2402.03460">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">stat.ML</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="Neural and Evolutionary Computing">cs.NE</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Combinatorics">math.CO</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"> Approximation Rates and VC-Dimension Bounds for (P)ReLU MLP Mixture of Experts </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Kratsios%2C+A">Anastasis Kratsios</a>, <a href="/search/cs?searchtype=author&query=Borde%2C+H+S+d+O">Haitz S谩ez de Oc谩riz Borde</a>, <a href="/search/cs?searchtype=author&query=Furuya%2C+T">Takashi Furuya</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M+T">Marc T. Law</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="2402.03460v2-abstract-short" style="display: inline;"> Mixture-of-Experts (MoEs) can scale up beyond traditional deep learning models by employing a routing strategy in which each input is processed by a single "expert" deep learning model. This strategy allows us to scale up the number of parameters defining the MoE while maintaining sparse activation, i.e., MoEs only load a small number of their total parameters into GPU VRAM for the forward pass de… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2402.03460v2-abstract-full').style.display = 'inline'; document.getElementById('2402.03460v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2402.03460v2-abstract-full" style="display: none;"> Mixture-of-Experts (MoEs) can scale up beyond traditional deep learning models by employing a routing strategy in which each input is processed by a single "expert" deep learning model. This strategy allows us to scale up the number of parameters defining the MoE while maintaining sparse activation, i.e., MoEs only load a small number of their total parameters into GPU VRAM for the forward pass depending on the input. In this paper, we provide an approximation and learning-theoretic analysis of mixtures of expert MLPs with (P)ReLU activation functions. We first prove that for every error level $\varepsilon>0$ and every Lipschitz function $f:[0,1]^n\to \mathbb{R}$, one can construct a MoMLP model (a Mixture-of-Experts comprising of (P)ReLU MLPs) which uniformly approximates $f$ to $\varepsilon$ accuracy over $[0,1]^n$, while only requiring networks of $\mathcal{O}(\varepsilon^{-1})$ parameters to be loaded in memory. Additionally, we show that MoMLPs can generalize since the entire MoMLP model has a (finite) VC dimension of $\tilde{O}(L\max\{nL,JW\})$, if there are $L$ experts and each expert has a depth and width of $J$ and $W$, respectively. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2402.03460v2-abstract-full').style.display = 'none'; document.getElementById('2402.03460v2-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 May, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 5 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/2402.01889">arXiv:2402.01889</a> <span> [<a href="https://arxiv.org/pdf/2402.01889">pdf</a>, <a href="https://arxiv.org/format/2402.01889">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> <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"> The Role of Foundation Models in Neuro-Symbolic Learning and Reasoning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Cunnington%2C+D">Daniel Cunnington</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M">Mark Law</a>, <a href="/search/cs?searchtype=author&query=Lobo%2C+J">Jorge Lobo</a>, <a href="/search/cs?searchtype=author&query=Russo%2C+A">Alessandra Russo</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="2402.01889v1-abstract-short" style="display: inline;"> Neuro-Symbolic AI (NeSy) holds promise to ensure the safe deployment of AI systems, as interpretable symbolic techniques provide formal behaviour guarantees. The challenge is how to effectively integrate neural and symbolic computation, to enable learning and reasoning from raw data. Existing pipelines that train the neural and symbolic components sequentially require extensive labelling, whereas… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2402.01889v1-abstract-full').style.display = 'inline'; document.getElementById('2402.01889v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2402.01889v1-abstract-full" style="display: none;"> Neuro-Symbolic AI (NeSy) holds promise to ensure the safe deployment of AI systems, as interpretable symbolic techniques provide formal behaviour guarantees. The challenge is how to effectively integrate neural and symbolic computation, to enable learning and reasoning from raw data. Existing pipelines that train the neural and symbolic components sequentially require extensive labelling, whereas end-to-end approaches are limited in terms of scalability, due to the combinatorial explosion in the symbol grounding problem. In this paper, we leverage the implicit knowledge within foundation models to enhance the performance in NeSy tasks, whilst reducing the amount of data labelling and manual engineering. We introduce a new architecture, called NeSyGPT, which fine-tunes a vision-language foundation model to extract symbolic features from raw data, before learning a highly expressive answer set program to solve a downstream task. Our comprehensive evaluation demonstrates that NeSyGPT has superior accuracy over various baselines, and can scale to complex NeSy tasks. Finally, we highlight the effective use of a large language model to generate the programmatic interface between the neural and symbolic components, significantly reducing the amount of manual engineering required. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2402.01889v1-abstract-full').style.display = 'none'; document.getElementById('2402.01889v1-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 February, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 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">Pre-print</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2312.04501">arXiv:2312.04501</a> <span> [<a href="https://arxiv.org/pdf/2312.04501">pdf</a>, <a href="https://arxiv.org/format/2312.04501">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="Machine Learning">stat.ML</span> </div> </div> <p class="title is-5 mathjax"> Graph Metanetworks for Processing Diverse Neural Architectures </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lim%2C+D">Derek Lim</a>, <a href="/search/cs?searchtype=author&query=Maron%2C+H">Haggai Maron</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M+T">Marc T. Law</a>, <a href="/search/cs?searchtype=author&query=Lorraine%2C+J">Jonathan Lorraine</a>, <a href="/search/cs?searchtype=author&query=Lucas%2C+J">James Lucas</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="2312.04501v2-abstract-short" style="display: inline;"> Neural networks efficiently encode learned information within their parameters. Consequently, many tasks can be unified by treating neural networks themselves as input data. When doing so, recent studies demonstrated the importance of accounting for the symmetries and geometry of parameter spaces. However, those works developed architectures tailored to specific networks such as MLPs and CNNs with… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.04501v2-abstract-full').style.display = 'inline'; document.getElementById('2312.04501v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2312.04501v2-abstract-full" style="display: none;"> Neural networks efficiently encode learned information within their parameters. Consequently, many tasks can be unified by treating neural networks themselves as input data. When doing so, recent studies demonstrated the importance of accounting for the symmetries and geometry of parameter spaces. However, those works developed architectures tailored to specific networks such as MLPs and CNNs without normalization layers, and generalizing such architectures to other types of networks can be challenging. In this work, we overcome these challenges by building new metanetworks - neural networks that take weights from other neural networks as input. Put simply, we carefully build graphs representing the input neural networks and process the graphs using graph neural networks. Our approach, Graph Metanetworks (GMNs), generalizes to neural architectures where competing methods struggle, such as multi-head attention layers, normalization layers, convolutional layers, ResNet blocks, and group-equivariant linear layers. We prove that GMNs are expressive and equivariant to parameter permutation symmetries that leave the input neural network functions unchanged. We validate the effectiveness of our method on several metanetwork tasks over diverse neural network architectures. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.04501v2-abstract-full').style.display = 'none'; document.getElementById('2312.04501v2-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">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">29 pages. v2 updated experimental results and details</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2310.12309">arXiv:2310.12309</a> <span> [<a href="https://arxiv.org/pdf/2310.12309">pdf</a>, <a href="https://arxiv.org/format/2310.12309">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> <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"> A Unifying Framework for Learning Argumentation Semantics </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Mileva%2C+Z">Zlatina Mileva</a>, <a href="/search/cs?searchtype=author&query=Bikakis%2C+A">Antonis Bikakis</a>, <a href="/search/cs?searchtype=author&query=D%27Asaro%2C+F+A">Fabio Aurelio D'Asaro</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M">Mark Law</a>, <a href="/search/cs?searchtype=author&query=Russo%2C+A">Alessandra Russo</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="2310.12309v1-abstract-short" style="display: inline;"> Argumentation is a very active research field of Artificial Intelligence concerned with the representation and evaluation of arguments used in dialogues between humans and/or artificial agents. Acceptability semantics of formal argumentation systems define the criteria for the acceptance or rejection of arguments. Several software systems, known as argumentation solvers, have been developed to com… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.12309v1-abstract-full').style.display = 'inline'; document.getElementById('2310.12309v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2310.12309v1-abstract-full" style="display: none;"> Argumentation is a very active research field of Artificial Intelligence concerned with the representation and evaluation of arguments used in dialogues between humans and/or artificial agents. Acceptability semantics of formal argumentation systems define the criteria for the acceptance or rejection of arguments. Several software systems, known as argumentation solvers, have been developed to compute the accepted/rejected arguments using such criteria. These include systems that learn to identify the accepted arguments using non-interpretable methods. In this paper we present a novel framework, which uses an Inductive Logic Programming approach to learn the acceptability semantics for several abstract and structured argumentation frameworks in an interpretable way. Through an empirical evaluation we show that our framework outperforms existing argumentation solvers, thus opening up new future research directions in the area of formal argumentation and human-machine dialogues. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.12309v1-abstract-full').style.display = 'none'; document.getElementById('2310.12309v1-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> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2309.10083">arXiv:2309.10083</a> <span> [<a href="https://arxiv.org/pdf/2309.10083">pdf</a>, <a href="https://arxiv.org/format/2309.10083">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Methodology">stat.ME</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"> Invariant Probabilistic Prediction </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Henzi%2C+A">Alexander Henzi</a>, <a href="/search/cs?searchtype=author&query=Shen%2C+X">Xinwei Shen</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M">Michael Law</a>, <a href="/search/cs?searchtype=author&query=B%C3%BChlmann%2C+P">Peter B眉hlmann</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="2309.10083v2-abstract-short" style="display: inline;"> In recent years, there has been a growing interest in statistical methods that exhibit robust performance under distribution changes between training and test data. While most of the related research focuses on point predictions with the squared error loss, this article turns the focus towards probabilistic predictions, which aim to comprehensively quantify the uncertainty of an outcome variable g… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2309.10083v2-abstract-full').style.display = 'inline'; document.getElementById('2309.10083v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2309.10083v2-abstract-full" style="display: none;"> In recent years, there has been a growing interest in statistical methods that exhibit robust performance under distribution changes between training and test data. While most of the related research focuses on point predictions with the squared error loss, this article turns the focus towards probabilistic predictions, which aim to comprehensively quantify the uncertainty of an outcome variable given covariates. Within a causality-inspired framework, we investigate the invariance and robustness of probabilistic predictions with respect to proper scoring rules. We show that arbitrary distribution shifts do not, in general, admit invariant and robust probabilistic predictions, in contrast to the setting of point prediction. We illustrate how to choose evaluation metrics and restrict the class of distribution shifts to allow for identifiability and invariance in the prototypical Gaussian heteroscedastic linear model. Motivated by these findings, we propose a method to yield invariant probabilistic predictions, called IPP, and study the consistency of the underlying parameters. Finally, we demonstrate the empirical performance of our proposed procedure on simulated as well as on single-cell data. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2309.10083v2-abstract-full').style.display = 'none'; document.getElementById('2309.10083v2-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 June, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 18 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/2302.04832">arXiv:2302.04832</a> <span> [<a href="https://arxiv.org/pdf/2302.04832">pdf</a>, <a href="https://arxiv.org/format/2302.04832">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"> Bridging the Sim2Real gap with CARE: Supervised Detection Adaptation with Conditional Alignment and Reweighting </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Prabhu%2C+V">Viraj Prabhu</a>, <a href="/search/cs?searchtype=author&query=Acuna%2C+D">David Acuna</a>, <a href="/search/cs?searchtype=author&query=Liao%2C+A">Andrew Liao</a>, <a href="/search/cs?searchtype=author&query=Mahmood%2C+R">Rafid Mahmood</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M+T">Marc T. Law</a>, <a href="/search/cs?searchtype=author&query=Hoffman%2C+J">Judy Hoffman</a>, <a href="/search/cs?searchtype=author&query=Fidler%2C+S">Sanja Fidler</a>, <a href="/search/cs?searchtype=author&query=Lucas%2C+J">James Lucas</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="2302.04832v1-abstract-short" style="display: inline;"> Sim2Real domain adaptation (DA) research focuses on the constrained setting of adapting from a labeled synthetic source domain to an unlabeled or sparsely labeled real target domain. However, for high-stakes applications (e.g. autonomous driving), it is common to have a modest amount of human-labeled real data in addition to plentiful auto-labeled source data (e.g. from a driving simulator). We st… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2302.04832v1-abstract-full').style.display = 'inline'; document.getElementById('2302.04832v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2302.04832v1-abstract-full" style="display: none;"> Sim2Real domain adaptation (DA) research focuses on the constrained setting of adapting from a labeled synthetic source domain to an unlabeled or sparsely labeled real target domain. However, for high-stakes applications (e.g. autonomous driving), it is common to have a modest amount of human-labeled real data in addition to plentiful auto-labeled source data (e.g. from a driving simulator). We study this setting of supervised sim2real DA applied to 2D object detection. We propose Domain Translation via Conditional Alignment and Reweighting (CARE) a novel algorithm that systematically exploits target labels to explicitly close the sim2real appearance and content gaps. We present an analytical justification of our algorithm and demonstrate strong gains over competing methods on standard benchmarks. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2302.04832v1-abstract-full').style.display = 'none'; document.getElementById('2302.04832v1-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> 9 February, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2210.01234">arXiv:2210.01234</a> <span> [<a href="https://arxiv.org/pdf/2210.01234">pdf</a>, <a href="https://arxiv.org/format/2210.01234">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="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> Optimizing Data Collection for Machine Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Mahmood%2C+R">Rafid Mahmood</a>, <a href="/search/cs?searchtype=author&query=Lucas%2C+J">James Lucas</a>, <a href="/search/cs?searchtype=author&query=Alvarez%2C+J+M">Jose M. Alvarez</a>, <a href="/search/cs?searchtype=author&query=Fidler%2C+S">Sanja Fidler</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M+T">Marc T. Law</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="2210.01234v1-abstract-short" style="display: inline;"> Modern deep learning systems require huge data sets to achieve impressive performance, but there is little guidance on how much or what kind of data to collect. Over-collecting data incurs unnecessary present costs, while under-collecting may incur future costs and delay workflows. We propose a new paradigm for modeling the data collection workflow as a formal optimal data collection problem that… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2210.01234v1-abstract-full').style.display = 'inline'; document.getElementById('2210.01234v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2210.01234v1-abstract-full" style="display: none;"> Modern deep learning systems require huge data sets to achieve impressive performance, but there is little guidance on how much or what kind of data to collect. Over-collecting data incurs unnecessary present costs, while under-collecting may incur future costs and delay workflows. We propose a new paradigm for modeling the data collection workflow as a formal optimal data collection problem that allows designers to specify performance targets, collection costs, a time horizon, and penalties for failing to meet the targets. Additionally, this formulation generalizes to tasks requiring multiple data sources, such as labeled and unlabeled data used in semi-supervised learning. To solve our problem, we develop Learn-Optimize-Collect (LOC), which minimizes expected future collection costs. Finally, we numerically compare our framework to the conventional baseline of estimating data requirements by extrapolating from neural scaling laws. We significantly reduce the risks of failing to meet desired performance targets on several classification, segmentation, and detection tasks, while maintaining low total collection costs. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2210.01234v1-abstract-full').style.display = 'none'; document.getElementById('2210.01234v1-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, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2022. </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 2022</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2207.01725">arXiv:2207.01725</a> <span> [<a href="https://arxiv.org/pdf/2207.01725">pdf</a>, <a href="https://arxiv.org/format/2207.01725">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> </div> </div> <p class="title is-5 mathjax"> How Much More Data Do I Need? Estimating Requirements for Downstream Tasks </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Mahmood%2C+R">Rafid Mahmood</a>, <a href="/search/cs?searchtype=author&query=Lucas%2C+J">James Lucas</a>, <a href="/search/cs?searchtype=author&query=Acuna%2C+D">David Acuna</a>, <a href="/search/cs?searchtype=author&query=Li%2C+D">Daiqing Li</a>, <a href="/search/cs?searchtype=author&query=Philion%2C+J">Jonah Philion</a>, <a href="/search/cs?searchtype=author&query=Alvarez%2C+J+M">Jose M. Alvarez</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+Z">Zhiding Yu</a>, <a href="/search/cs?searchtype=author&query=Fidler%2C+S">Sanja Fidler</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M+T">Marc T. Law</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="2207.01725v2-abstract-short" style="display: inline;"> Given a small training data set and a learning algorithm, how much more data is necessary to reach a target validation or test performance? This question is of critical importance in applications such as autonomous driving or medical imaging where collecting data is expensive and time-consuming. Overestimating or underestimating data requirements incurs substantial costs that could be avoided with… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2207.01725v2-abstract-full').style.display = 'inline'; document.getElementById('2207.01725v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2207.01725v2-abstract-full" style="display: none;"> Given a small training data set and a learning algorithm, how much more data is necessary to reach a target validation or test performance? This question is of critical importance in applications such as autonomous driving or medical imaging where collecting data is expensive and time-consuming. Overestimating or underestimating data requirements incurs substantial costs that could be avoided with an adequate budget. Prior work on neural scaling laws suggest that the power-law function can fit the validation performance curve and extrapolate it to larger data set sizes. We find that this does not immediately translate to the more difficult downstream task of estimating the required data set size to meet a target performance. In this work, we consider a broad class of computer vision tasks and systematically investigate a family of functions that generalize the power-law function to allow for better estimation of data requirements. Finally, we show that incorporating a tuned correction factor and collecting over multiple rounds significantly improves the performance of the data estimators. Using our guidelines, practitioners can accurately estimate data requirements of machine learning systems to gain savings in both development time and data acquisition costs. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2207.01725v2-abstract-full').style.display = 'none'; document.getElementById('2207.01725v2-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 July, 2022; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 4 July, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2022. </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 CVPR 2022</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2205.15752">arXiv:2205.15752</a> <span> [<a href="https://arxiv.org/pdf/2205.15752">pdf</a>, <a href="https://arxiv.org/format/2205.15752">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"> Hierarchies of Reward Machines </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Furelos-Blanco%2C+D">Daniel Furelos-Blanco</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M">Mark Law</a>, <a href="/search/cs?searchtype=author&query=Jonsson%2C+A">Anders Jonsson</a>, <a href="/search/cs?searchtype=author&query=Broda%2C+K">Krysia Broda</a>, <a href="/search/cs?searchtype=author&query=Russo%2C+A">Alessandra Russo</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="2205.15752v2-abstract-short" style="display: inline;"> Reward machines (RMs) are a recent formalism for representing the reward function of a reinforcement learning task through a finite-state machine whose edges encode subgoals of the task using high-level events. The structure of RMs enables the decomposition of a task into simpler and independently solvable subtasks that help tackle long-horizon and/or sparse reward tasks. We propose a formalism fo… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2205.15752v2-abstract-full').style.display = 'inline'; document.getElementById('2205.15752v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2205.15752v2-abstract-full" style="display: none;"> Reward machines (RMs) are a recent formalism for representing the reward function of a reinforcement learning task through a finite-state machine whose edges encode subgoals of the task using high-level events. The structure of RMs enables the decomposition of a task into simpler and independently solvable subtasks that help tackle long-horizon and/or sparse reward tasks. We propose a formalism for further abstracting the subtask structure by endowing an RM with the ability to call other RMs, thus composing a hierarchy of RMs (HRM). We exploit HRMs by treating each call to an RM as an independently solvable subtask using the options framework, and describe a curriculum-based method to learn HRMs from traces observed by the agent. Our experiments reveal that exploiting a handcrafted HRM leads to faster convergence than with a flat HRM, and that learning an HRM is feasible in cases where its equivalent flat representation is not. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2205.15752v2-abstract-full').style.display = 'none'; document.getElementById('2205.15752v2-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 June, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 31 May, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 2022. </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">Preprint accepted for publication to the 40th International Conference on Machine Learning (ICML-23)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2205.12735">arXiv:2205.12735</a> <span> [<a href="https://arxiv.org/pdf/2205.12735">pdf</a>, <a href="https://arxiv.org/format/2205.12735">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"> Neuro-Symbolic Learning of Answer Set Programs from Raw Data </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Cunnington%2C+D">Daniel Cunnington</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M">Mark Law</a>, <a href="/search/cs?searchtype=author&query=Lobo%2C+J">Jorge Lobo</a>, <a href="/search/cs?searchtype=author&query=Russo%2C+A">Alessandra Russo</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="2205.12735v8-abstract-short" style="display: inline;"> One of the ultimate goals of Artificial Intelligence is to assist humans in complex decision making. A promising direction for achieving this goal is Neuro-Symbolic AI, which aims to combine the interpretability of symbolic techniques with the ability of deep learning to learn from raw data. However, most current approaches require manually engineered symbolic knowledge, and where end-to-end train… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2205.12735v8-abstract-full').style.display = 'inline'; document.getElementById('2205.12735v8-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2205.12735v8-abstract-full" style="display: none;"> One of the ultimate goals of Artificial Intelligence is to assist humans in complex decision making. A promising direction for achieving this goal is Neuro-Symbolic AI, which aims to combine the interpretability of symbolic techniques with the ability of deep learning to learn from raw data. However, most current approaches require manually engineered symbolic knowledge, and where end-to-end training is considered, such approaches are either restricted to learning definite programs, or are restricted to training binary neural networks. In this paper, we introduce Neuro-Symbolic Inductive Learner (NSIL), an approach that trains a general neural network to extract latent concepts from raw data, whilst learning symbolic knowledge that maps latent concepts to target labels. The novelty of our approach is a method for biasing the learning of symbolic knowledge, based on the in-training performance of both neural and symbolic components. We evaluate NSIL on three problem domains of different complexity, including an NP-complete problem. Our results demonstrate that NSIL learns expressive knowledge, solves computationally complex problems, and achieves state-of-the-art performance in terms of accuracy and data efficiency. Code and technical appendix: https://github.com/DanCunnington/NSIL <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2205.12735v8-abstract-full').style.display = 'none'; document.getElementById('2205.12735v8-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 February, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 25 May, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 2022. </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 IJCAI 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/2205.07129">arXiv:2205.07129</a> <span> [<a href="https://arxiv.org/pdf/2205.07129">pdf</a>, <a href="https://arxiv.org/format/2205.07129">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> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Logic in Computer Science">cs.LO</span> </div> </div> <p class="title is-5 mathjax"> Efficient lifting of symmetry breaking constraints for complex combinatorial problems </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Tarzariol%2C+A">Alice Tarzariol</a>, <a href="/search/cs?searchtype=author&query=Gebser%2C+M">Martin Gebser</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M">Mark Law</a>, <a href="/search/cs?searchtype=author&query=Schekotihin%2C+K">Konstantin Schekotihin</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="2205.07129v1-abstract-short" style="display: inline;"> Many industrial applications require finding solutions to challenging combinatorial problems. Efficient elimination of symmetric solution candidates is one of the key enablers for high-performance solving. However, existing model-based approaches for symmetry breaking are limited to problems for which a set of representative and easily-solvable instances is available, which is often not the case i… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2205.07129v1-abstract-full').style.display = 'inline'; document.getElementById('2205.07129v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2205.07129v1-abstract-full" style="display: none;"> Many industrial applications require finding solutions to challenging combinatorial problems. Efficient elimination of symmetric solution candidates is one of the key enablers for high-performance solving. However, existing model-based approaches for symmetry breaking are limited to problems for which a set of representative and easily-solvable instances is available, which is often not the case in practical applications. This work extends the learning framework and implementation of a model-based approach for Answer Set Programming to overcome these limitations and address challenging problems, such as the Partner Units Problem. In particular, we incorporate a new conflict analysis algorithm in the Inductive Logic Programming system ILASP, redefine the learning task, and suggest a new example generation method to scale up the approach. The experiments conducted for different kinds of Partner Units Problem instances demonstrate the applicability of our approach and the computational benefits due to the first-order constraints learned. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2205.07129v1-abstract-full').style.display = 'none'; document.getElementById('2205.07129v1-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 May, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 2022. </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">Paper presented at the 38th International Conference on Logic Programming (ICLP 2022), 16 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/2202.05352">arXiv:2202.05352</a> <span> [<a href="https://arxiv.org/pdf/2202.05352">pdf</a>, <a href="https://arxiv.org/format/2202.05352">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="Computer Vision and Pattern Recognition">cs.CV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computer Science and Game Theory">cs.GT</span> </div> </div> <p class="title is-5 mathjax"> Domain Adversarial Training: A Game Perspective </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Acuna%2C+D">David Acuna</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M+T">Marc T Law</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+G">Guojun Zhang</a>, <a href="/search/cs?searchtype=author&query=Fidler%2C+S">Sanja Fidler</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="2202.05352v1-abstract-short" style="display: inline;"> The dominant line of work in domain adaptation has focused on learning invariant representations using domain-adversarial training. In this paper, we interpret this approach from a game theoretical perspective. Defining optimal solutions in domain-adversarial training as a local Nash equilibrium, we show that gradient descent in domain-adversarial training can violate the asymptotic convergence gu… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2202.05352v1-abstract-full').style.display = 'inline'; document.getElementById('2202.05352v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2202.05352v1-abstract-full" style="display: none;"> The dominant line of work in domain adaptation has focused on learning invariant representations using domain-adversarial training. In this paper, we interpret this approach from a game theoretical perspective. Defining optimal solutions in domain-adversarial training as a local Nash equilibrium, we show that gradient descent in domain-adversarial training can violate the asymptotic convergence guarantees of the optimizer, oftentimes hindering the transfer performance. Our analysis leads us to replace gradient descent with high-order ODE solvers (i.e., Runge-Kutta), for which we derive asymptotic convergence guarantees. This family of optimizers is significantly more stable and allows more aggressive learning rates, leading to high performance gains when used as a drop-in replacement over standard optimizers. Our experiments show that in conjunction with state-of-the-art domain-adversarial methods, we achieve up to 3.5% improvement with less than of half training iterations. Our optimizers are easy to implement, free of additional parameters, and can be plugged into any domain-adversarial framework. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2202.05352v1-abstract-full').style.display = 'none'; document.getElementById('2202.05352v1-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 February, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2022. </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">ICLR 2022</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2111.05956">arXiv:2111.05956</a> <span> [<a href="https://arxiv.org/pdf/2111.05956">pdf</a>, <a href="https://arxiv.org/format/2111.05956">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> </div> </div> <p class="title is-5 mathjax"> Feature Generation for Long-tail Classification </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Vigneswaran%2C+R">Rahul Vigneswaran</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M+T">Marc T. Law</a>, <a href="/search/cs?searchtype=author&query=Balasubramanian%2C+V+N">Vineeth N. Balasubramanian</a>, <a href="/search/cs?searchtype=author&query=Tapaswi%2C+M">Makarand Tapaswi</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="2111.05956v1-abstract-short" style="display: inline;"> The visual world naturally exhibits an imbalance in the number of object or scene instances resulting in a \emph{long-tailed distribution}. This imbalance poses significant challenges for classification models based on deep learning. Oversampling instances of the tail classes attempts to solve this imbalance. However, the limited visual diversity results in a network with poor representation abili… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2111.05956v1-abstract-full').style.display = 'inline'; document.getElementById('2111.05956v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2111.05956v1-abstract-full" style="display: none;"> The visual world naturally exhibits an imbalance in the number of object or scene instances resulting in a \emph{long-tailed distribution}. This imbalance poses significant challenges for classification models based on deep learning. Oversampling instances of the tail classes attempts to solve this imbalance. However, the limited visual diversity results in a network with poor representation ability. A simple counter to this is decoupling the representation and classifier networks and using oversampling only to train the classifier. In this paper, instead of repeatedly re-sampling the same image (and thereby features), we explore a direction that attempts to generate meaningful features by estimating the tail category's distribution. Inspired by ideas from recent work on few-shot learning, we create calibrated distributions to sample additional features that are subsequently used to train the classifier. Through several experiments on the CIFAR-100-LT (long-tail) dataset with varying imbalance factors and on mini-ImageNet-LT (long-tail), we show the efficacy of our approach and establish a new state-of-the-art. We also present a qualitative analysis of generated features using t-SNE visualizations and analyze the nearest neighbors used to calibrate the tail class distributions. Our code is available at https://github.com/rahulvigneswaran/TailCalibX. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2111.05956v1-abstract-full').style.display = 'none'; document.getElementById('2111.05956v1-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 November, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2021. </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 ICVGIP'21. Code available at https://github.com/rahulvigneswaran/TailCalibX</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2106.13103">arXiv:2106.13103</a> <span> [<a href="https://arxiv.org/pdf/2106.13103">pdf</a>, <a href="https://arxiv.org/format/2106.13103">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> </div> </div> <p class="title is-5 mathjax"> FF-NSL: Feed-Forward Neural-Symbolic Learner </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Cunnington%2C+D">Daniel Cunnington</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M">Mark Law</a>, <a href="/search/cs?searchtype=author&query=Russo%2C+A">Alessandra Russo</a>, <a href="/search/cs?searchtype=author&query=Lobo%2C+J">Jorge Lobo</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="2106.13103v3-abstract-short" style="display: inline;"> Logic-based machine learning aims to learn general, interpretable knowledge in a data-efficient manner. However, labelled data must be specified in a structured logical form. To address this limitation, we propose a neural-symbolic learning framework, called Feed-Forward Neural-Symbolic Learner (FFNSL), that integrates a logic-based machine learning system capable of learning from noisy examples,… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2106.13103v3-abstract-full').style.display = 'inline'; document.getElementById('2106.13103v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2106.13103v3-abstract-full" style="display: none;"> Logic-based machine learning aims to learn general, interpretable knowledge in a data-efficient manner. However, labelled data must be specified in a structured logical form. To address this limitation, we propose a neural-symbolic learning framework, called Feed-Forward Neural-Symbolic Learner (FFNSL), that integrates a logic-based machine learning system capable of learning from noisy examples, with neural networks, in order to learn interpretable knowledge from labelled unstructured data. We demonstrate the generality of FFNSL on four neural-symbolic classification problems, where different pre-trained neural network models and logic-based machine learning systems are integrated to learn interpretable knowledge from sequences of images. We evaluate the robustness of our framework by using images subject to distributional shifts, for which the pre-trained neural networks may predict incorrectly and with high confidence. We analyse the impact that these shifts have on the accuracy of the learned knowledge and run-time performance, comparing FFNSL to tree-based and pure neural approaches. Our experimental results show that FFNSL outperforms the baselines by learning more accurate and interpretable knowledge with fewer examples. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2106.13103v3-abstract-full').style.display = 'none'; document.getElementById('2106.13103v3-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 January, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 24 June, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2021. </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">Pre-print, work in progress</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2106.11344">arXiv:2106.11344</a> <span> [<a href="https://arxiv.org/pdf/2106.11344">pdf</a>, <a href="https://arxiv.org/format/2106.11344">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="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> f-Domain-Adversarial Learning: Theory and Algorithms </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Acuna%2C+D">David Acuna</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+G">Guojun Zhang</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M+T">Marc T. Law</a>, <a href="/search/cs?searchtype=author&query=Fidler%2C+S">Sanja Fidler</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="2106.11344v1-abstract-short" style="display: inline;"> Unsupervised domain adaptation is used in many machine learning applications where, during training, a model has access to unlabeled data in the target domain, and a related labeled dataset. In this paper, we introduce a novel and general domain-adversarial framework. Specifically, we derive a novel generalization bound for domain adaptation that exploits a new measure of discrepancy between distr… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2106.11344v1-abstract-full').style.display = 'inline'; document.getElementById('2106.11344v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2106.11344v1-abstract-full" style="display: none;"> Unsupervised domain adaptation is used in many machine learning applications where, during training, a model has access to unlabeled data in the target domain, and a related labeled dataset. In this paper, we introduce a novel and general domain-adversarial framework. Specifically, we derive a novel generalization bound for domain adaptation that exploits a new measure of discrepancy between distributions based on a variational characterization of f-divergences. It recovers the theoretical results from Ben-David et al. (2010a) as a special case and supports divergences used in practice. Based on this bound, we derive a new algorithmic framework that introduces a key correction in the original adversarial training method of Ganin et al. (2016). We show that many regularizers and ad-hoc objectives introduced over the last years in this framework are then not required to achieve performance comparable to (if not better than) state-of-the-art domain-adversarial methods. Experimental analysis conducted on real-world natural language and computer vision datasets show that our framework outperforms existing baselines, and obtains the best results for f-divergences that were not considered previously in domain-adversarial learning. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2106.11344v1-abstract-full').style.display = 'none'; document.getElementById('2106.11344v1-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 June, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2021. </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 2021</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2106.02968">arXiv:2106.02968</a> <span> [<a href="https://arxiv.org/pdf/2106.02968">pdf</a>, <a href="https://arxiv.org/format/2106.02968">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="Optimization and Control">math.OC</span> </div> </div> <p class="title is-5 mathjax"> Low Budget Active Learning via Wasserstein Distance: An Integer Programming Approach </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Mahmood%2C+R">Rafid Mahmood</a>, <a href="/search/cs?searchtype=author&query=Fidler%2C+S">Sanja Fidler</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M+T">Marc T. Law</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="2106.02968v4-abstract-short" style="display: inline;"> Active learning is the process of training a model with limited labeled data by selecting a core subset of an unlabeled data pool to label. The large scale of data sets used in deep learning forces most sample selection strategies to employ efficient heuristics. This paper introduces an integer optimization problem for selecting a core set that minimizes the discrete Wasserstein distance from the… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2106.02968v4-abstract-full').style.display = 'inline'; document.getElementById('2106.02968v4-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2106.02968v4-abstract-full" style="display: none;"> Active learning is the process of training a model with limited labeled data by selecting a core subset of an unlabeled data pool to label. The large scale of data sets used in deep learning forces most sample selection strategies to employ efficient heuristics. This paper introduces an integer optimization problem for selecting a core set that minimizes the discrete Wasserstein distance from the unlabeled pool. We demonstrate that this problem can be tractably solved with a Generalized Benders Decomposition algorithm. Our strategy uses high-quality latent features that can be obtained by unsupervised learning on the unlabeled pool. Numerical results on several data sets show that our optimization approach is competitive with baselines and particularly outperforms them in the low budget regime where less than one percent of the data set is labeled. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2106.02968v4-abstract-full').style.display = 'none'; document.getElementById('2106.02968v4-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 March, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 5 June, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2021. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2101.00058">arXiv:2101.00058</a> <span> [<a href="https://arxiv.org/pdf/2101.00058">pdf</a>, <a href="https://arxiv.org/ps/2101.00058">ps</a>, <a href="https://arxiv.org/format/2101.00058">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"> Conflict-driven Inductive Logic Programming </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Law%2C+M">Mark Law</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="2101.00058v3-abstract-short" style="display: inline;"> The goal of Inductive Logic Programming (ILP) is to learn a program that explains a set of examples. Until recently, most research on ILP targeted learning Prolog programs. The ILASP system instead learns Answer Set Programs (ASP). Learning such expressive programs widens the applicability of ILP considerably; for example, enabling preference learning, learning common-sense knowledge, including de… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2101.00058v3-abstract-full').style.display = 'inline'; document.getElementById('2101.00058v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2101.00058v3-abstract-full" style="display: none;"> The goal of Inductive Logic Programming (ILP) is to learn a program that explains a set of examples. Until recently, most research on ILP targeted learning Prolog programs. The ILASP system instead learns Answer Set Programs (ASP). Learning such expressive programs widens the applicability of ILP considerably; for example, enabling preference learning, learning common-sense knowledge, including defaults and exceptions, and learning non-deterministic theories. Early versions of ILASP can be considered meta-level ILP approaches, which encode a learning task as a logic program and delegate the search to an ASP solver. More recently, ILASP has shifted towards a new method, inspired by conflict-driven SAT and ASP solvers. The fundamental idea of the approach, called Conflict-driven ILP (CDILP), is to iteratively interleave the search for a hypothesis with the generation of constraints which explain why the current hypothesis does not cover a particular example. These coverage constraints allow ILASP to rule out not just the current hypothesis, but an entire class of hypotheses that do not satisfy the coverage constraint. This paper formalises the CDILP approach and presents the ILASP3 and ILASP4 systems for CDILP, which are demonstrated to be more scalable than previous ILASP systems, particularly in the presence of noise. Under consideration in Theory and Practice of Logic Programming (TPLP). <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2101.00058v3-abstract-full').style.display = 'none'; document.getElementById('2101.00058v3-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 January, 2022; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 31 December, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> January 2021. </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">Under consideration in Theory and Practice of Logic Programming (TPLP)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2012.05023">arXiv:2012.05023</a> <span> [<a href="https://arxiv.org/pdf/2012.05023">pdf</a>, <a href="https://arxiv.org/ps/2012.05023">ps</a>, <a href="https://arxiv.org/format/2012.05023">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"> NSL: Hybrid Interpretable Learning From Noisy Raw Data </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Cunnington%2C+D">Daniel Cunnington</a>, <a href="/search/cs?searchtype=author&query=Russo%2C+A">Alessandra Russo</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M">Mark Law</a>, <a href="/search/cs?searchtype=author&query=Lobo%2C+J">Jorge Lobo</a>, <a href="/search/cs?searchtype=author&query=Kaplan%2C+L">Lance Kaplan</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="2012.05023v2-abstract-short" style="display: inline;"> Inductive Logic Programming (ILP) systems learn generalised, interpretable rules in a data-efficient manner utilising existing background knowledge. However, current ILP systems require training examples to be specified in a structured logical format. Neural networks learn from unstructured data, although their learned models may be difficult to interpret and are vulnerable to data perturbations a… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2012.05023v2-abstract-full').style.display = 'inline'; document.getElementById('2012.05023v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2012.05023v2-abstract-full" style="display: none;"> Inductive Logic Programming (ILP) systems learn generalised, interpretable rules in a data-efficient manner utilising existing background knowledge. However, current ILP systems require training examples to be specified in a structured logical format. Neural networks learn from unstructured data, although their learned models may be difficult to interpret and are vulnerable to data perturbations at run-time. This paper introduces a hybrid neural-symbolic learning framework, called NSL, that learns interpretable rules from labelled unstructured data. NSL combines pre-trained neural networks for feature extraction with FastLAS, a state-of-the-art ILP system for rule learning under the answer set semantics. Features extracted by the neural components define the structured context of labelled examples and the confidence of the neural predictions determines the level of noise of the examples. Using the scoring function of FastLAS, NSL searches for short, interpretable rules that generalise over such noisy examples. We evaluate our framework on propositional and first-order classification tasks using the MNIST dataset as raw data. Specifically, we demonstrate that NSL is able to learn robust rules from perturbed MNIST data and achieve comparable or superior accuracy when compared to neural network and random forest baselines whilst being more general and interpretable. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2012.05023v2-abstract-full').style.display = 'none'; document.getElementById('2012.05023v2-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 June, 2021; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 9 December, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2020. </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 article has been replaced with arXiv:2106.13103</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2011.14488">arXiv:2011.14488</a> <span> [<a href="https://arxiv.org/pdf/2011.14488">pdf</a>, <a href="https://arxiv.org/format/2011.14488">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"> Self-Supervised Real-to-Sim Scene Generation </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Prakash%2C+A">Aayush Prakash</a>, <a href="/search/cs?searchtype=author&query=Debnath%2C+S">Shoubhik Debnath</a>, <a href="/search/cs?searchtype=author&query=Lafleche%2C+J">Jean-Francois Lafleche</a>, <a href="/search/cs?searchtype=author&query=Cameracci%2C+E">Eric Cameracci</a>, <a href="/search/cs?searchtype=author&query=State%2C+G">Gavriel State</a>, <a href="/search/cs?searchtype=author&query=Birchfield%2C+S">Stan Birchfield</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M+T">Marc T. Law</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="2011.14488v2-abstract-short" style="display: inline;"> Synthetic data is emerging as a promising solution to the scalability issue of supervised deep learning, especially when real data are difficult to acquire or hard to annotate. Synthetic data generation, however, can itself be prohibitively expensive when domain experts have to manually and painstakingly oversee the process. Moreover, neural networks trained on synthetic data often do not perform… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2011.14488v2-abstract-full').style.display = 'inline'; document.getElementById('2011.14488v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2011.14488v2-abstract-full" style="display: none;"> Synthetic data is emerging as a promising solution to the scalability issue of supervised deep learning, especially when real data are difficult to acquire or hard to annotate. Synthetic data generation, however, can itself be prohibitively expensive when domain experts have to manually and painstakingly oversee the process. Moreover, neural networks trained on synthetic data often do not perform well on real data because of the domain gap. To solve these challenges, we propose Sim2SG, a self-supervised automatic scene generation technique for matching the distribution of real data. Importantly, Sim2SG does not require supervision from the real-world dataset, thus making it applicable in situations for which such annotations are difficult to obtain. Sim2SG is designed to bridge both the content and appearance gaps, by matching the content of real data, and by matching the features in the source and target domains. We select scene graph (SG) generation as the downstream task, due to the limited availability of labeled datasets. Experiments demonstrate significant improvements over leading baselines in reducing the domain gap both qualitatively and quantitatively, on several synthetic datasets as well as the real-world KITTI dataset. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2011.14488v2-abstract-full').style.display = 'none'; document.getElementById('2011.14488v2-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 August, 2021; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 29 November, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2020. </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 ICCV 2021. Project page: https://research.nvidia.com/publication/2021-08_Sim2SG</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2009.03855">arXiv:2009.03855</a> <span> [<a href="https://arxiv.org/pdf/2009.03855">pdf</a>, <a href="https://arxiv.org/format/2009.03855">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> <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/10.1613/jair.1.12372">10.1613/jair.1.12372 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Induction and Exploitation of Subgoal Automata for Reinforcement Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Furelos-Blanco%2C+D">Daniel Furelos-Blanco</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M">Mark Law</a>, <a href="/search/cs?searchtype=author&query=Jonsson%2C+A">Anders Jonsson</a>, <a href="/search/cs?searchtype=author&query=Broda%2C+K">Krysia Broda</a>, <a href="/search/cs?searchtype=author&query=Russo%2C+A">Alessandra Russo</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="2009.03855v2-abstract-short" style="display: inline;"> In this paper we present ISA, an approach for learning and exploiting subgoals in episodic reinforcement learning (RL) tasks. ISA interleaves reinforcement learning with the induction of a subgoal automaton, an automaton whose edges are labeled by the task's subgoals expressed as propositional logic formulas over a set of high-level events. A subgoal automaton also consists of two special states:… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2009.03855v2-abstract-full').style.display = 'inline'; document.getElementById('2009.03855v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2009.03855v2-abstract-full" style="display: none;"> In this paper we present ISA, an approach for learning and exploiting subgoals in episodic reinforcement learning (RL) tasks. ISA interleaves reinforcement learning with the induction of a subgoal automaton, an automaton whose edges are labeled by the task's subgoals expressed as propositional logic formulas over a set of high-level events. A subgoal automaton also consists of two special states: a state indicating the successful completion of the task, and a state indicating that the task has finished without succeeding. A state-of-the-art inductive logic programming system is used to learn a subgoal automaton that covers the traces of high-level events observed by the RL agent. When the currently exploited automaton does not correctly recognize a trace, the automaton learner induces a new automaton that covers that trace. The interleaving process guarantees the induction of automata with the minimum number of states, and applies a symmetry breaking mechanism to shrink the search space whilst remaining complete. We evaluate ISA in several gridworld and continuous state space problems using different RL algorithms that leverage the automaton structures. We provide an in-depth empirical analysis of the automaton learning performance in terms of the traces, the symmetry breaking and specific restrictions imposed on the final learnable automaton. For each class of RL problem, we show that the learned automata can be successfully exploited to learn policies that reach the goal, achieving an average reward comparable to the case where automata are not learned but handcrafted and given beforehand. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2009.03855v2-abstract-full').style.display = 'none'; document.getElementById('2009.03855v2-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 March, 2021; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 8 September, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 2020. </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 in the Journal of Artificial Intelligence Research (JAIR)</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> Journal of Artificial Intelligence Research, 70, 1031-1116 (2021) </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2007.03643">arXiv:2007.03643</a> <span> [<a href="https://arxiv.org/pdf/2007.03643">pdf</a>, <a href="https://arxiv.org/format/2007.03643">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"> Segmentation of Pulmonary Opacification in Chest CT Scans of COVID-19 Patients </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lensink%2C+K">Keegan Lensink</a>, <a href="/search/cs?searchtype=author&query=Laradji%2C+I">Issam Laradji</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M">Marco Law</a>, <a href="/search/cs?searchtype=author&query=Barbano%2C+P+E">Paolo Emilio Barbano</a>, <a href="/search/cs?searchtype=author&query=Nicolaou%2C+S">Savvas Nicolaou</a>, <a href="/search/cs?searchtype=author&query=Parker%2C+W">William Parker</a>, <a href="/search/cs?searchtype=author&query=Haber%2C+E">Eldad Haber</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="2007.03643v2-abstract-short" style="display: inline;"> The Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) has rapidly spread into a global pandemic. A form of pneumonia, presenting as opacities with in a patient's lungs, is the most common presentation associated with this virus, and great attention has gone into how these changes relate to patient morbidity and mortality. In this work we provide open source models for the segmentation o… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2007.03643v2-abstract-full').style.display = 'inline'; document.getElementById('2007.03643v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2007.03643v2-abstract-full" style="display: none;"> The Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-CoV-2) has rapidly spread into a global pandemic. A form of pneumonia, presenting as opacities with in a patient's lungs, is the most common presentation associated with this virus, and great attention has gone into how these changes relate to patient morbidity and mortality. In this work we provide open source models for the segmentation of patterns of pulmonary opacification on chest Computed Tomography (CT) scans which have been correlated with various stages and severities of infection. We have collected 663 chest CT scans of COVID-19 patients from healthcare centers around the world, and created pixel wise segmentation labels for nearly 25,000 slices that segment 6 different patterns of pulmonary opacification. We provide open source implementations and pre-trained weights for multiple segmentation models trained on our dataset. Our best model achieves an opacity Intersection-Over-Union score of 0.76 on our test set, demonstrates successful domain adaptation, and predicts the volume of opacification within 1.7\% of expert radiologists. Additionally, we present an analysis of the inter-observer variability inherent to this task, and propose methods for appropriate probabilistic approaches. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2007.03643v2-abstract-full').style.display = 'none'; document.getElementById('2007.03643v2-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, 2020; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 7 July, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2020. </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">9 pages, 5 figures. Fix typo in delimiter between author names in arXiv metadata</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2007.02180">arXiv:2007.02180</a> <span> [<a href="https://arxiv.org/pdf/2007.02180">pdf</a>, <a href="https://arxiv.org/format/2007.02180">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"> A Weakly Supervised Consistency-based Learning Method for COVID-19 Segmentation in CT Images </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Laradji%2C+I">Issam Laradji</a>, <a href="/search/cs?searchtype=author&query=Rodriguez%2C+P">Pau Rodriguez</a>, <a href="/search/cs?searchtype=author&query=Ma%C3%B1as%2C+O">Oscar Ma帽as</a>, <a href="/search/cs?searchtype=author&query=Lensink%2C+K">Keegan Lensink</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M">Marco Law</a>, <a href="/search/cs?searchtype=author&query=Kurzman%2C+L">Lironne Kurzman</a>, <a href="/search/cs?searchtype=author&query=Parker%2C+W">William Parker</a>, <a href="/search/cs?searchtype=author&query=Vazquez%2C+D">David Vazquez</a>, <a href="/search/cs?searchtype=author&query=Nowrouzezahrai%2C+D">Derek Nowrouzezahrai</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="2007.02180v2-abstract-short" style="display: inline;"> Coronavirus Disease 2019 (COVID-19) has spread aggressively across the world causing an existential health crisis. Thus, having a system that automatically detects COVID-19 in tomography (CT) images can assist in quantifying the severity of the illness. Unfortunately, labelling chest CT scans requires significant domain expertise, time, and effort. We address these labelling challenges by only req… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2007.02180v2-abstract-full').style.display = 'inline'; document.getElementById('2007.02180v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2007.02180v2-abstract-full" style="display: none;"> Coronavirus Disease 2019 (COVID-19) has spread aggressively across the world causing an existential health crisis. Thus, having a system that automatically detects COVID-19 in tomography (CT) images can assist in quantifying the severity of the illness. Unfortunately, labelling chest CT scans requires significant domain expertise, time, and effort. We address these labelling challenges by only requiring point annotations, a single pixel for each infected region on a CT image. This labeling scheme allows annotators to label a pixel in a likely infected region, only taking 1-3 seconds, as opposed to 10-15 seconds to segment a region. Conventionally, segmentation models train on point-level annotations using the cross-entropy loss function on these labels. However, these models often suffer from low precision. Thus, we propose a consistency-based (CB) loss function that encourages the output predictions to be consistent with spatial transformations of the input images. The experiments on 3 open-source COVID-19 datasets show that this loss function yields significant improvement over conventional point-level loss functions and almost matches the performance of models trained with full supervision with much less human effort. Code is available at: \url{https://github.com/IssamLaradji/covid19_weak_supervision}. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2007.02180v2-abstract-full').style.display = 'none'; document.getElementById('2007.02180v2-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, 2020; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 4 July, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2020. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2007.00211">arXiv:2007.00211</a> <span> [<a href="https://arxiv.org/pdf/2007.00211">pdf</a>, <a href="https://arxiv.org/format/2007.00211">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="Machine Learning">stat.ML</span> </div> </div> <p class="title is-5 mathjax"> Ultrahyperbolic Representation Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Law%2C+M+T">Marc T. Law</a>, <a href="/search/cs?searchtype=author&query=Stam%2C+J">Jos Stam</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="2007.00211v5-abstract-short" style="display: inline;"> In machine learning, data is usually represented in a (flat) Euclidean space where distances between points are along straight lines. Researchers have recently considered more exotic (non-Euclidean) Riemannian manifolds such as hyperbolic space which is well suited for tree-like data. In this paper, we propose a representation living on a pseudo-Riemannian manifold of constant nonzero curvature. I… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2007.00211v5-abstract-full').style.display = 'inline'; document.getElementById('2007.00211v5-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2007.00211v5-abstract-full" style="display: none;"> In machine learning, data is usually represented in a (flat) Euclidean space where distances between points are along straight lines. Researchers have recently considered more exotic (non-Euclidean) Riemannian manifolds such as hyperbolic space which is well suited for tree-like data. In this paper, we propose a representation living on a pseudo-Riemannian manifold of constant nonzero curvature. It is a generalization of hyperbolic and spherical geometries where the nondegenerate metric tensor need not be positive definite. We provide the necessary learning tools in this geometry and extend gradient-based optimization techniques. More specifically, we provide closed-form expressions for distances via geodesics and define a descent direction to minimize some objective function. Our novel framework is applied to graph representations. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2007.00211v5-abstract-full').style.display = 'none'; document.getElementById('2007.00211v5-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 January, 2021; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 30 June, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2020. </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 2020</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2005.00904">arXiv:2005.00904</a> <span> [<a href="https://arxiv.org/pdf/2005.00904">pdf</a>, <a href="https://arxiv.org/ps/2005.00904">ps</a>, <a href="https://arxiv.org/format/2005.00904">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> <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"> The ILASP system for Inductive Learning of Answer Set Programs </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Law%2C+M">Mark Law</a>, <a href="/search/cs?searchtype=author&query=Russo%2C+A">Alessandra Russo</a>, <a href="/search/cs?searchtype=author&query=Broda%2C+K">Krysia Broda</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="2005.00904v1-abstract-short" style="display: inline;"> The goal of Inductive Logic Programming (ILP) is to learn a program that explains a set of examples in the context of some pre-existing background knowledge. Until recently, most research on ILP targeted learning Prolog programs. Our own ILASP system instead learns Answer Set Programs, including normal rules, choice rules and hard and weak constraints. Learning such expressive programs widens the… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2005.00904v1-abstract-full').style.display = 'inline'; document.getElementById('2005.00904v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2005.00904v1-abstract-full" style="display: none;"> The goal of Inductive Logic Programming (ILP) is to learn a program that explains a set of examples in the context of some pre-existing background knowledge. Until recently, most research on ILP targeted learning Prolog programs. Our own ILASP system instead learns Answer Set Programs, including normal rules, choice rules and hard and weak constraints. Learning such expressive programs widens the applicability of ILP considerably; for example, enabling preference learning, learning common-sense knowledge, including defaults and exceptions, and learning non-deterministic theories. In this paper, we first give a general overview of ILASP's learning framework and its capabilities. This is followed by a comprehensive summary of the evolution of the ILASP system, presenting the strengths and weaknesses of each version, with a particular emphasis on scalability. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2005.00904v1-abstract-full').style.display = 'none'; document.getElementById('2005.00904v1-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 May, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 2020. </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 the ALP newsletter</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1911.13152">arXiv:1911.13152</a> <span> [<a href="https://arxiv.org/pdf/1911.13152">pdf</a>, <a href="https://arxiv.org/format/1911.13152">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="Logic in Computer Science">cs.LO</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"> Induction of Subgoal Automata for Reinforcement Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Furelos-Blanco%2C+D">Daniel Furelos-Blanco</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M">Mark Law</a>, <a href="/search/cs?searchtype=author&query=Russo%2C+A">Alessandra Russo</a>, <a href="/search/cs?searchtype=author&query=Broda%2C+K">Krysia Broda</a>, <a href="/search/cs?searchtype=author&query=Jonsson%2C+A">Anders Jonsson</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="1911.13152v1-abstract-short" style="display: inline;"> In this work we present ISA, a novel approach for learning and exploiting subgoals in reinforcement learning (RL). Our method relies on inducing an automaton whose transitions are subgoals expressed as propositional formulas over a set of observable events. A state-of-the-art inductive logic programming system is used to learn the automaton from observation traces perceived by the RL agent. The re… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1911.13152v1-abstract-full').style.display = 'inline'; document.getElementById('1911.13152v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1911.13152v1-abstract-full" style="display: none;"> In this work we present ISA, a novel approach for learning and exploiting subgoals in reinforcement learning (RL). Our method relies on inducing an automaton whose transitions are subgoals expressed as propositional formulas over a set of observable events. A state-of-the-art inductive logic programming system is used to learn the automaton from observation traces perceived by the RL agent. The reinforcement learning and automaton learning processes are interleaved: a new refined automaton is learned whenever the RL agent generates a trace not recognized by the current automaton. We evaluate ISA in several gridworld problems and show that it performs similarly to a method for which automata are given in advance. We also show that the learned automata can be exploited to speed up convergence through reward shaping and transfer learning across multiple tasks. Finally, we analyze the running time and the number of traces that ISA needs to learn an automata, and the impact that the number of observable events has on the learner's performance. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1911.13152v1-abstract-full').style.display = 'none'; document.getElementById('1911.13152v1-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, 2019; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2019. </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">Preprint accepted for publication to the 34th AAAI Conference on Artificial Intelligence (AAAI-20)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1910.04961">arXiv:1910.04961</a> <span> [<a href="https://arxiv.org/pdf/1910.04961">pdf</a>, <a href="https://arxiv.org/format/1910.04961">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/10.1007/978-3-030-32226-7_84">10.1007/978-3-030-32226-7_84 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Adversarial Pulmonary Pathology Translation for Pairwise Chest X-ray Data Augmentation </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Xing%2C+Y">Yunyan Xing</a>, <a href="/search/cs?searchtype=author&query=Ge%2C+Z">Zongyuan Ge</a>, <a href="/search/cs?searchtype=author&query=Zeng%2C+R">Rui Zeng</a>, <a href="/search/cs?searchtype=author&query=Mahapatra%2C+D">Dwarikanath Mahapatra</a>, <a href="/search/cs?searchtype=author&query=Seah%2C+J">Jarrel Seah</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M">Meng Law</a>, <a href="/search/cs?searchtype=author&query=Drummond%2C+T">Tom Drummond</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="1910.04961v2-abstract-short" style="display: inline;"> Recent works show that Generative Adversarial Networks (GANs) can be successfully applied to chest X-ray data augmentation for lung disease recognition. However, the implausible and distorted pathology features generated from the less than perfect generator may lead to wrong clinical decisions. Why not keep the original pathology region? We proposed a novel approach that allows our generative mode… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1910.04961v2-abstract-full').style.display = 'inline'; document.getElementById('1910.04961v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1910.04961v2-abstract-full" style="display: none;"> Recent works show that Generative Adversarial Networks (GANs) can be successfully applied to chest X-ray data augmentation for lung disease recognition. However, the implausible and distorted pathology features generated from the less than perfect generator may lead to wrong clinical decisions. Why not keep the original pathology region? We proposed a novel approach that allows our generative model to generate high quality plausible images that contain undistorted pathology areas. The main idea is to design a training scheme based on an image-to-image translation network to introduce variations of new lung features around the pathology ground-truth area. Moreover, our model is able to leverage both annotated disease images and unannotated healthy lung images for the purpose of generation. We demonstrate the effectiveness of our model on two tasks: (i) we invite certified radiologists to assess the quality of the generated synthetic images against real and other state-of-the-art generative models, and (ii) data augmentation to improve the performance of disease localisation. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1910.04961v2-abstract-full').style.display = 'none'; document.getElementById('1910.04961v2-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 January, 2020; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 10 October, 2019; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2019. </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">Code: https://github.com/yunyanxing/pairwise_xray_augmentation - Accepted to the International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) 2019</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1909.11722">arXiv:1909.11722</a> <span> [<a href="https://arxiv.org/pdf/1909.11722">pdf</a>, <a href="https://arxiv.org/format/1909.11722">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="Machine Learning">stat.ML</span> </div> </div> <p class="title is-5 mathjax"> A Theoretical Analysis of the Number of Shots in Few-Shot Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Cao%2C+T">Tianshi Cao</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M">Marc Law</a>, <a href="/search/cs?searchtype=author&query=Fidler%2C+S">Sanja Fidler</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="1909.11722v2-abstract-short" style="display: inline;"> Few-shot classification is the task of predicting the category of an example from a set of few labeled examples. The number of labeled examples per category is called the number of shots (or shot number). Recent works tackle this task through meta-learning, where a meta-learner extracts information from observed tasks during meta-training to quickly adapt to new tasks during meta-testing. In this… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1909.11722v2-abstract-full').style.display = 'inline'; document.getElementById('1909.11722v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1909.11722v2-abstract-full" style="display: none;"> Few-shot classification is the task of predicting the category of an example from a set of few labeled examples. The number of labeled examples per category is called the number of shots (or shot number). Recent works tackle this task through meta-learning, where a meta-learner extracts information from observed tasks during meta-training to quickly adapt to new tasks during meta-testing. In this formulation, the number of shots exploited during meta-training has an impact on the recognition performance at meta-test time. Generally, the shot number used in meta-training should match the one used in meta-testing to obtain the best performance. We introduce a theoretical analysis of the impact of the shot number on Prototypical Networks, a state-of-the-art few-shot classification method. From our analysis, we propose a simple method that is robust to the choice of shot number used during meta-training, which is a crucial hyperparameter. The performance of our model trained for an arbitrary meta-training shot number shows great performance for different values of meta-testing shot numbers. We experimentally demonstrate our approach on different few-shot classification benchmarks. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1909.11722v2-abstract-full').style.display = 'none'; document.getElementById('1909.11722v2-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 February, 2020; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 25 September, 2019; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 2019. </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 incl. appendix, 6 figures</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> 8th International Conference on Learning Representations 2020, Addis Ababa, Ethiopia </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1908.03381">arXiv:1908.03381</a> <span> [<a href="https://arxiv.org/pdf/1908.03381">pdf</a>, <a href="https://arxiv.org/format/1908.03381">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"> Video Face Clustering with Unknown Number of Clusters </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Tapaswi%2C+M">Makarand Tapaswi</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M+T">Marc T. Law</a>, <a href="/search/cs?searchtype=author&query=Fidler%2C+S">Sanja Fidler</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="1908.03381v2-abstract-short" style="display: inline;"> Understanding videos such as TV series and movies requires analyzing who the characters are and what they are doing. We address the challenging problem of clustering face tracks based on their identity. Different from previous work in this area, we choose to operate in a realistic and difficult setting where: (i) the number of characters is not known a priori; and (ii) face tracks belonging to min… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1908.03381v2-abstract-full').style.display = 'inline'; document.getElementById('1908.03381v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1908.03381v2-abstract-full" style="display: none;"> Understanding videos such as TV series and movies requires analyzing who the characters are and what they are doing. We address the challenging problem of clustering face tracks based on their identity. Different from previous work in this area, we choose to operate in a realistic and difficult setting where: (i) the number of characters is not known a priori; and (ii) face tracks belonging to minor or background characters are not discarded. To this end, we propose Ball Cluster Learning (BCL), a supervised approach to carve the embedding space into balls of equal size, one for each cluster. The learned ball radius is easily translated to a stopping criterion for iterative merging algorithms. This gives BCL the ability to estimate the number of clusters as well as their assignment, achieving promising results on commonly used datasets. We also present a thorough discussion of how existing metric learning literature can be adapted for this task. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1908.03381v2-abstract-full').style.display = 'none'; document.getElementById('1908.03381v2-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 August, 2019; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 9 August, 2019; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> August 2019. </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 2019, code and data at https://github.com/makarandtapaswi/BallClustering_ICCV2019</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1906.09627">arXiv:1906.09627</a> <span> [<a href="https://arxiv.org/pdf/1906.09627">pdf</a>, <a href="https://arxiv.org/format/1906.09627">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> <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"> Inductive general game playing </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Cropper%2C+A">Andrew Cropper</a>, <a href="/search/cs?searchtype=author&query=Evans%2C+R">Richard Evans</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M">Mark Law</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="1906.09627v1-abstract-short" style="display: inline;"> General game playing (GGP) is a framework for evaluating an agent's general intelligence across a wide range of tasks. In the GGP competition, an agent is given the rules of a game (described as a logic program) that it has never seen before. The task is for the agent to play the game, thus generating game traces. The winner of the GGP competition is the agent that gets the best total score over a… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1906.09627v1-abstract-full').style.display = 'inline'; document.getElementById('1906.09627v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1906.09627v1-abstract-full" style="display: none;"> General game playing (GGP) is a framework for evaluating an agent's general intelligence across a wide range of tasks. In the GGP competition, an agent is given the rules of a game (described as a logic program) that it has never seen before. The task is for the agent to play the game, thus generating game traces. The winner of the GGP competition is the agent that gets the best total score over all the games. In this paper, we invert this task: a learner is given game traces and the task is to learn the rules that could produce the traces. This problem is central to inductive general game playing (IGGP). We introduce a technique that automatically generates IGGP tasks from GGP games. We introduce an IGGP dataset which contains traces from 50 diverse games, such as Sudoku, Sokoban, and Checkers. We claim that IGGP is difficult for existing inductive logic programming (ILP) approaches. To support this claim, we evaluate existing ILP systems on our dataset. Our empirical results show that most of the games cannot be correctly learned by existing systems. The best performing system solves only 40% of the tasks perfectly. Our results suggest that IGGP poses many challenges to existing approaches. Furthermore, because we can automatically generate IGGP tasks from GGP games, our dataset will continue to grow with the GGP competition, as new games are added every year. We therefore think that the IGGP problem and dataset will be valuable for motivating and evaluating future research. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1906.09627v1-abstract-full').style.display = 'none'; document.getElementById('1906.09627v1-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 June, 2019; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2019. </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 the Machine Learning journal</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1902.02375">arXiv:1902.02375</a> <span> [<a href="https://arxiv.org/pdf/1902.02375">pdf</a>, <a href="https://arxiv.org/format/1902.02375">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="Sound">cs.SD</span> <span class="tag is-small is-grey 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="Machine Learning">stat.ML</span> </div> </div> <p class="title is-5 mathjax"> Centroid-based deep metric learning for speaker recognition </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Wang%2C+J">Jixuan Wang</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+K">Kuan-Chieh Wang</a>, <a href="/search/cs?searchtype=author&query=Law%2C+M">Marc Law</a>, <a href="/search/cs?searchtype=author&query=Rudzicz%2C+F">Frank Rudzicz</a>, <a href="/search/cs?searchtype=author&query=Brudno%2C+M">Michael Brudno</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="1902.02375v1-abstract-short" style="display: inline;"> Speaker embedding models that utilize neural networks to map utterances to a space where distances reflect similarity between speakers have driven recent progress in the speaker recognition task. However, there is still a significant performance gap between recognizing speakers in the training set and unseen speakers. The latter case corresponds to the few-shot learning task, where a trained model… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1902.02375v1-abstract-full').style.display = 'inline'; document.getElementById('1902.02375v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1902.02375v1-abstract-full" style="display: none;"> Speaker embedding models that utilize neural networks to map utterances to a space where distances reflect similarity between speakers have driven recent progress in the speaker recognition task. However, there is still a significant performance gap between recognizing speakers in the training set and unseen speakers. The latter case corresponds to the few-shot learning task, where a trained model is evaluated on unseen classes. Here, we optimize a speaker embedding model with prototypical network loss (PNL), a state-of-the-art approach for the few-shot image classification task. The resulting embedding model outperforms the state-of-the-art triplet loss based models in both speaker verification and identification tasks, for both seen and unseen speakers. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1902.02375v1-abstract-full').style.display = 'none'; document.getElementById('1902.02375v1-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 February, 2019; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2019. </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">ICASSP 2019 (44th International Conference on Acoustics, Speech, and Signal Processing)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1808.08441">arXiv:1808.08441</a> <span> [<a href="https://arxiv.org/pdf/1808.08441">pdf</a>, <a href="https://arxiv.org/ps/1808.08441">ps</a>, <a href="https://arxiv.org/format/1808.08441">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"> Inductive Learning of Answer Set Programs from Noisy Examples </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Law%2C+M">Mark Law</a>, <a href="/search/cs?searchtype=author&query=Russo%2C+A">Alessandra Russo</a>, <a href="/search/cs?searchtype=author&query=Broda%2C+K">Krysia Broda</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="1808.08441v1-abstract-short" style="display: inline;"> In recent years, non-monotonic Inductive Logic Programming has received growing interest. Specifically, several new learning frameworks and algorithms have been introduced for learning under the answer set semantics, allowing the learning of common-sense knowledge involving defaults and exceptions, which are essential aspects of human reasoning. In this paper, we present a noise-tolerant generalis… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1808.08441v1-abstract-full').style.display = 'inline'; document.getElementById('1808.08441v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1808.08441v1-abstract-full" style="display: none;"> In recent years, non-monotonic Inductive Logic Programming has received growing interest. Specifically, several new learning frameworks and algorithms have been introduced for learning under the answer set semantics, allowing the learning of common-sense knowledge involving defaults and exceptions, which are essential aspects of human reasoning. In this paper, we present a noise-tolerant generalisation of the learning from answer sets framework. We evaluate our ILASP3 system, both on synthetic and on real datasets, represented in the new framework. In particular, we show that on many of the datasets ILASP3 achieves a higher accuracy than other ILP systems that have previously been applied to the datasets, including a recently proposed differentiable learning framework. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1808.08441v1-abstract-full').style.display = 'none'; document.getElementById('1808.08441v1-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 August, 2018; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> August 2018. </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">To appear in Advances in Cognitive Systems</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1608.01946">arXiv:1608.01946</a> <span> [<a href="https://arxiv.org/pdf/1608.01946">pdf</a>, <a href="https://arxiv.org/format/1608.01946">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"> Iterative Learning of Answer Set Programs from Context Dependent Examples </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Law%2C+M">Mark Law</a>, <a href="/search/cs?searchtype=author&query=Russo%2C+A">Alessandra Russo</a>, <a href="/search/cs?searchtype=author&query=Broda%2C+K">Krysia Broda</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="1608.01946v1-abstract-short" style="display: inline;"> In recent years, several frameworks and systems have been proposed that extend Inductive Logic Programming (ILP) to the Answer Set Programming (ASP) paradigm. In ILP, examples must all be explained by a hypothesis together with a given background knowledge. In existing systems, the background knowledge is the same for all examples; however, examples may be context-dependent. This means that some e… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1608.01946v1-abstract-full').style.display = 'inline'; document.getElementById('1608.01946v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1608.01946v1-abstract-full" style="display: none;"> In recent years, several frameworks and systems have been proposed that extend Inductive Logic Programming (ILP) to the Answer Set Programming (ASP) paradigm. In ILP, examples must all be explained by a hypothesis together with a given background knowledge. In existing systems, the background knowledge is the same for all examples; however, examples may be context-dependent. This means that some examples should be explained in the context of some information, whereas others should be explained in different contexts. In this paper, we capture this notion and present a context-dependent extension of the Learning from Ordered Answer Sets framework. In this extension, contexts can be used to further structure the background knowledge. We then propose a new iterative algorithm, ILASP2i, which exploits this feature to scale up the existing ILASP2 system to learning tasks with large numbers of examples. We demonstrate the gain in scalability by applying both algorithms to various learning tasks. Our results show that, compared to ILASP2, the newly proposed ILASP2i system can be two orders of magnitude faster and use two orders of magnitude less memory, whilst preserving the same average accuracy. This paper is under consideration for acceptance in TPLP. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1608.01946v1-abstract-full').style.display = 'none'; document.getElementById('1608.01946v1-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 August, 2016; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> August 2016. </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">Paper presented at the 32nd International Conference on Logic Programming (ICLP 2016), New York City, USA, 16-21 October 2016, 22 pages, LaTeX, 3 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/1507.06566">arXiv:1507.06566</a> <span> [<a href="https://arxiv.org/pdf/1507.06566">pdf</a>, <a href="https://arxiv.org/ps/1507.06566">ps</a>, <a href="https://arxiv.org/format/1507.06566">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 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/10.1017/S1471068415000198">10.1017/S1471068415000198 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Learning Weak Constraints in Answer Set Programming </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Law%2C+M">Mark Law</a>, <a href="/search/cs?searchtype=author&query=Russo%2C+A">Alessandra Russo</a>, <a href="/search/cs?searchtype=author&query=Broda%2C+K">Krysia Broda</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="1507.06566v1-abstract-short" style="display: inline;"> This paper contributes to the area of inductive logic programming by presenting a new learning framework that allows the learning of weak constraints in Answer Set Programming (ASP). The framework, called Learning from Ordered Answer Sets, generalises our previous work on learning ASP programs without weak constraints, by considering a new notion of examples as ordered pairs of partial answer sets… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1507.06566v1-abstract-full').style.display = 'inline'; document.getElementById('1507.06566v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1507.06566v1-abstract-full" style="display: none;"> This paper contributes to the area of inductive logic programming by presenting a new learning framework that allows the learning of weak constraints in Answer Set Programming (ASP). The framework, called Learning from Ordered Answer Sets, generalises our previous work on learning ASP programs without weak constraints, by considering a new notion of examples as ordered pairs of partial answer sets that exemplify which answer sets of a learned hypothesis (together with a given background knowledge) are preferred to others. In this new learning task inductive solutions are searched within a hypothesis space of normal rules, choice rules, and hard and weak constraints. We propose a new algorithm, ILASP2, which is sound and complete with respect to our new learning framework. We investigate its applicability to learning preferences in an interview scheduling problem and also demonstrate that when restricted to the task of learning ASP programs without weak constraints, ILASP2 can be much more efficient than our previously proposed system. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1507.06566v1-abstract-full').style.display = 'none'; document.getElementById('1507.06566v1-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 July, 2015; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2015. </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">To appear in Theory and Practice of Logic Programming (TPLP), Proceedings of ICLP 2015</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> Theory and Practice of Logic Programming 15 (2015) 511-525 </p> </li> </ol> <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