Search | arXiv e-print repository

<!DOCTYPE html> <html lang="en"> <head> <meta charset="utf-8"/> <meta name="viewport" content="width=device-width, initial-scale=1"/>  <link rel="apple-touch-icon" sizes="180x180" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/apple-touch-icon.png"> <link rel="icon" type="image/png" sizes="32x32" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/favicon-32x32.png"> <link rel="icon" type="image/png" sizes="16x16" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/favicon-16x16.png"> <link rel="manifest" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/site.webmanifest"> <link rel="mask-icon" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/safari-pinned-tab.svg" color="#b31b1b"> <link rel="shortcut icon" href="https://static.arxiv.org/static/base/1.0.0a5/images/icons/favicon.ico"> <meta name="msapplication-TileColor" content="#b31b1b"> <meta name="msapplication-config" content="images/icons/browserconfig.xml"> <meta name="theme-color" content="#b31b1b">  <title>Search | arXiv e-print repository</title> <script defer src="https://static.arxiv.org/static/base/1.0.0a5/fontawesome-free-5.11.2-web/js/all.js"></script> <link rel="stylesheet" href="https://static.arxiv.org/static/base/1.0.0a5/css/arxivstyle.css" /> <script type="text/x-mathjax-config"> MathJax.Hub.Config({ messageStyle: "none", extensions: ["tex2jax.js"], jax: ["input/TeX", "output/HTML-CSS"], tex2jax: { inlineMath: [ ['$','$'], ["\$","\$"] ], displayMath: [ ['$$','$$'], ["\\[","\\]"] ], processEscapes: true, ignoreClass: '.*', processClass: 'mathjax.*' }, TeX: { extensions: ["AMSmath.js", "AMSsymbols.js", "noErrors.js"], noErrors: { inlineDelimiters: ["$","$"], multiLine: false, style: { "font-size": "normal", "border": "" } } }, "HTML-CSS": { availableFonts: ["TeX"] } }); </script> <script src='//static.arxiv.org/MathJax-2.7.3/MathJax.js'></script> <script src="https://static.arxiv.org/static/base/1.0.0a5/js/notification.js"></script> <link rel="stylesheet" href="https://static.arxiv.org/static/search/0.5.6/css/bulma-tooltip.min.css" /> <link rel="stylesheet" href="https://static.arxiv.org/static/search/0.5.6/css/search.css" /> <script src="https://code.jquery.com/jquery-3.2.1.slim.min.js" integrity="sha256-k2WSCIexGzOj3Euiig+TlR8gA0EmPjuc79OEeY5L45g=" crossorigin="anonymous"></script> <script src="https://static.arxiv.org/static/search/0.5.6/js/fieldset.js"></script> <style> radio#cf-customfield_11400 { display: none; } </style> </head> <body> <header><a href="#main-container" class="is-sr-only">Skip to main content</a>  <div class="attribution level is-marginless" role="banner"> <div class="level-left"> <a class="level-item" href="https://cornell.edu/"><img src="https://static.arxiv.org/static/base/1.0.0a5/images/cornell-reduced-white-SMALL.svg" alt="Cornell University" width="200" aria-label="logo" /></a> </div> <div class="level-right is-marginless"><p class="sponsors level-item is-marginless"><span id="support-ack-url">We gratefully acknowledge support from<br /> the Simons Foundation, <a href="https://info.arxiv.org/about/ourmembers.html">member institutions</a>, and all contributors. <a href="https://info.arxiv.org/about/donate.html">Donate</a></span></p></div> </div>  <div class="identity level is-marginless"> <div class="level-left"> <div class="level-item"> <a class="arxiv" href="https://arxiv.org/" aria-label="arxiv-logo"> <img src="https://static.arxiv.org/static/base/1.0.0a5/images/arxiv-logo-one-color-white.svg" aria-label="logo" alt="arxiv logo" width="85" style="width:85px;"/> </a> </div> </div> <div class="search-block level-right"> <form class="level-item mini-search" method="GET" action="https://arxiv.org/search"> <div class="field has-addons"> <div class="control"> <input class="input is-small" type="text" name="query" placeholder="Search..." aria-label="Search term or terms" /> <p class="help"><a href="https://info.arxiv.org/help">Help</a> | <a href="https://arxiv.org/search/advanced">Advanced Search</a></p> </div> <div class="control"> <div class="select is-small"> <select name="searchtype" aria-label="Field to search"> <option value="all" selected="selected">All fields</option> <option value="title">Title</option> <option value="author">Author</option> <option value="abstract">Abstract</option> <option value="comments">Comments</option> <option value="journal_ref">Journal reference</option> <option value="acm_class">ACM classification</option> <option value="msc_class">MSC classification</option> <option value="report_num">Report number</option> <option value="paper_id">arXiv identifier</option> <option value="doi">DOI</option> <option value="orcid">ORCID</option> <option value="author_id">arXiv author ID</option> <option value="help">Help pages</option> <option value="full_text">Full text</option> </select> </div> </div> <input type="hidden" name="source" value="header"> <button class="button is-small is-cul-darker">Search</button> </div> </form> </div> </div>  <div class="container"> <div class="user-tools is-size-7 has-text-right has-text-weight-bold" role="navigation" aria-label="User menu"> <a href="https://arxiv.org/login">Login</a> </div> </div> </header> <main class="container" id="main-container"> <div class="level is-marginless"> <div class="level-left"> <h1 class="title is-clearfix"> Showing 1–50 of 142 results for author: <span class="mathjax">Foster, I</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=Foster%2C+I">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="Foster, I"> </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=Foster%2C+I&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="Foster, I"> <ul id="abstracts"><li><input checked id="abstracts-0" name="abstracts" type="radio" value="show"> <label for="abstracts-0">Show abstracts</label></li><li><input id="abstracts-1" name="abstracts" type="radio" value="hide"> <label for="abstracts-1">Hide abstracts</label></li></ul> </div> <div class="box field is-grouped is-grouped-multiline level-item"> <div class="control"> <span class="select is-small"> <select id="size" name="size"><option value="25">25</option><option selected value="50">50</option><option value="100">100</option><option value="200">200</option></select> </span> <label for="size">results per page</label>. </div> <div class="control"> <label for="order">Sort results by</label> <span class="select is-small"> <select id="order" name="order"><option selected value="-announced_date_first">Announcement date (newest first)</option><option value="announced_date_first">Announcement date (oldest first)</option><option value="-submitted_date">Submission date (newest first)</option><option value="submitted_date">Submission date (oldest first)</option><option value="">Relevance</option></select> </span> </div> <div class="control"> <button class="button is-small is-link">Go</button> </div> </div> </form> </div> </div> <nav class="pagination is-small is-centered breathe-horizontal" role="navigation" aria-label="pagination"> <a href="" class="pagination-previous is-invisible">Previous </a> <a href="/search/?searchtype=author&query=Foster%2C+I&start=50" class="pagination-next" >Next </a> <ul class="pagination-list"> <li> <a href="/search/?searchtype=author&query=Foster%2C+I&start=0" class="pagination-link is-current" aria-label="Goto page 1">1 </a> </li> <li> <a href="/search/?searchtype=author&query=Foster%2C+I&start=50" class="pagination-link " aria-label="Page 2" aria-current="page">2 </a> </li> <li> <a href="/search/?searchtype=author&query=Foster%2C+I&start=100" class="pagination-link " aria-label="Page 3" aria-current="page">3 </a> </li> </ul> </nav> <ol class="breathe-horizontal" start="1"> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.15221">arXiv:2411.15221</a> <span> [<a href="https://arxiv.org/pdf/2411.15221">pdf</a>, <a href="https://arxiv.org/format/2411.15221">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="Materials Science">cond-mat.mtrl-sci</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Chemical Physics">physics.chem-ph</span> </div> </div> <p class="title is-5 mathjax"> Reflections from the 2024 Large Language Model (LLM) Hackathon for Applications in Materials Science and Chemistry </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zimmermann%2C+Y">Yoel Zimmermann</a>, <a href="/search/cs?searchtype=author&query=Bazgir%2C+A">Adib Bazgir</a>, <a href="/search/cs?searchtype=author&query=Afzal%2C+Z">Zartashia Afzal</a>, <a href="/search/cs?searchtype=author&query=Agbere%2C+F">Fariha Agbere</a>, <a href="/search/cs?searchtype=author&query=Ai%2C+Q">Qianxiang Ai</a>, <a href="/search/cs?searchtype=author&query=Alampara%2C+N">Nawaf Alampara</a>, <a href="/search/cs?searchtype=author&query=Al-Feghali%2C+A">Alexander Al-Feghali</a>, <a href="/search/cs?searchtype=author&query=Ansari%2C+M">Mehrad Ansari</a>, <a href="/search/cs?searchtype=author&query=Antypov%2C+D">Dmytro Antypov</a>, <a href="/search/cs?searchtype=author&query=Aswad%2C+A">Amro Aswad</a>, <a href="/search/cs?searchtype=author&query=Bai%2C+J">Jiaru Bai</a>, <a href="/search/cs?searchtype=author&query=Baibakova%2C+V">Viktoriia Baibakova</a>, <a href="/search/cs?searchtype=author&query=Biswajeet%2C+D+D">Devi Dutta Biswajeet</a>, <a href="/search/cs?searchtype=author&query=Bitzek%2C+E">Erik Bitzek</a>, <a href="/search/cs?searchtype=author&query=Bocarsly%2C+J+D">Joshua D. Bocarsly</a>, <a href="/search/cs?searchtype=author&query=Borisova%2C+A">Anna Borisova</a>, <a href="/search/cs?searchtype=author&query=Bran%2C+A+M">Andres M Bran</a>, <a href="/search/cs?searchtype=author&query=Brinson%2C+L+C">L. Catherine Brinson</a>, <a href="/search/cs?searchtype=author&query=Calderon%2C+M+M">Marcel Moran Calderon</a>, <a href="/search/cs?searchtype=author&query=Canalicchio%2C+A">Alessandro Canalicchio</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+V">Victor Chen</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+Y">Yuan Chiang</a>, <a href="/search/cs?searchtype=author&query=Circi%2C+D">Defne Circi</a>, <a href="/search/cs?searchtype=author&query=Charmes%2C+B">Benjamin Charmes</a>, <a href="/search/cs?searchtype=author&query=Chaudhary%2C+V">Vikrant Chaudhary</a> , et al. (116 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="2411.15221v1-abstract-short" style="display: inline;"> Here, we present the outcomes from the second Large Language Model (LLM) Hackathon for Applications in Materials Science and Chemistry, which engaged participants across global hybrid locations, resulting in 34 team submissions. The submissions spanned seven key application areas and demonstrated the diverse utility of LLMs for applications in (1) molecular and material property prediction; (2) mo… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.15221v1-abstract-full').style.display = 'inline'; document.getElementById('2411.15221v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.15221v1-abstract-full" style="display: none;"> Here, we present the outcomes from the second Large Language Model (LLM) Hackathon for Applications in Materials Science and Chemistry, which engaged participants across global hybrid locations, resulting in 34 team submissions. The submissions spanned seven key application areas and demonstrated the diverse utility of LLMs for applications in (1) molecular and material property prediction; (2) molecular and material design; (3) automation and novel interfaces; (4) scientific communication and education; (5) research data management and automation; (6) hypothesis generation and evaluation; and (7) knowledge extraction and reasoning from scientific literature. Each team submission is presented in a summary table with links to the code and as brief papers in the appendix. Beyond team results, we discuss the hackathon event and its hybrid format, which included physical hubs in Toronto, Montreal, San Francisco, Berlin, Lausanne, and Tokyo, alongside a global online hub to enable local and virtual collaboration. Overall, the event highlighted significant improvements in LLM capabilities since the previous year's hackathon, suggesting continued expansion of LLMs for applications in materials science and chemistry research. These outcomes demonstrate the dual utility of LLMs as both multipurpose models for diverse machine learning tasks and platforms for rapid prototyping custom applications in scientific research. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.15221v1-abstract-full').style.display = 'none'; document.getElementById('2411.15221v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 20 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">98 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/2411.04257">arXiv:2411.04257</a> <span> [<a href="https://arxiv.org/pdf/2411.04257">pdf</a>, <a href="https://arxiv.org/format/2411.04257">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"> LSHBloom: Memory-efficient, Extreme-scale Document Deduplication </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Khan%2C+A">Arham Khan</a>, <a href="/search/cs?searchtype=author&query=Underwood%2C+R">Robert Underwood</a>, <a href="/search/cs?searchtype=author&query=Siebenschuh%2C+C">Carlo Siebenschuh</a>, <a href="/search/cs?searchtype=author&query=Babuji%2C+Y">Yadu Babuji</a>, <a href="/search/cs?searchtype=author&query=Ajith%2C+A">Aswathy Ajith</a>, <a href="/search/cs?searchtype=author&query=Hippe%2C+K">Kyle Hippe</a>, <a href="/search/cs?searchtype=author&query=Gokdemir%2C+O">Ozan Gokdemir</a>, <a href="/search/cs?searchtype=author&query=Brace%2C+A">Alexander Brace</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</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="2411.04257v1-abstract-short" style="display: inline;"> Deduplication is a major focus for assembling and curating training datasets for large language models (LLM) -- detecting and eliminating additional instances of the same content -- in large collections of technical documents. Unrestrained, duplicates in the training dataset increase training costs and lead to undesirable properties such as memorization in trained models or cheating on evaluation.… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.04257v1-abstract-full').style.display = 'inline'; document.getElementById('2411.04257v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.04257v1-abstract-full" style="display: none;"> Deduplication is a major focus for assembling and curating training datasets for large language models (LLM) -- detecting and eliminating additional instances of the same content -- in large collections of technical documents. Unrestrained, duplicates in the training dataset increase training costs and lead to undesirable properties such as memorization in trained models or cheating on evaluation. Contemporary approaches to document-level deduplication are often extremely expensive in both runtime and memory. We propose LSHBloom, an extension to MinhashLSH, which replaces the expensive LSHIndex with lightweight Bloom filters. LSHBloom demonstrates the same deduplication performance as MinhashLSH with only a marginal increase in false positives (as low as 1e-5 in our experiments); demonstrates competitive runtime (270\% faster than MinhashLSH on peS2o); and, crucially, uses just 0.6\% of the disk space required by MinhashLSH to deduplicate peS2o. We demonstrate that this space advantage scales with increased dataset size -- at the extreme scale of several billion documents, LSHBloom promises a 250\% speedup and a 54$\times$ space advantage over traditional MinHashLSH scaling deduplication of text datasets to many billions of documents. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.04257v1-abstract-full').style.display = 'none'; document.getElementById('2411.04257v1-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 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/2410.14943">arXiv:2410.14943</a> <span> [<a href="https://arxiv.org/pdf/2410.14943">pdf</a>, <a href="https://arxiv.org/format/2410.14943">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Distributed, Parallel, and Cluster Computing">cs.DC</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/10.5281/zenodo.13844758">10.5281/zenodo.13844758 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Workflows Community Summit 2024: Future Trends and Challenges in Scientific Workflows </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=da+Silva%2C+R+F">Rafael Ferreira da Silva</a>, <a href="/search/cs?searchtype=author&query=Bard%2C+D">Deborah Bard</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=de+Witt%2C+S">Shaun de Witt</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I+T">Ian T. Foster</a>, <a href="/search/cs?searchtype=author&query=Gibbs%2C+T">Tom Gibbs</a>, <a href="/search/cs?searchtype=author&query=Goble%2C+C">Carole Goble</a>, <a href="/search/cs?searchtype=author&query=Godoy%2C+W">William Godoy</a>, <a href="/search/cs?searchtype=author&query=Gustafsson%2C+J">Johan Gustafsson</a>, <a href="/search/cs?searchtype=author&query=Haus%2C+U">Utz-Uwe Haus</a>, <a href="/search/cs?searchtype=author&query=Hudson%2C+S">Stephen Hudson</a>, <a href="/search/cs?searchtype=author&query=Jha%2C+S">Shantenu Jha</a>, <a href="/search/cs?searchtype=author&query=Los%2C+L">Laila Los</a>, <a href="/search/cs?searchtype=author&query=Paine%2C+D">Drew Paine</a>, <a href="/search/cs?searchtype=author&query=Suter%2C+F">Fr茅d茅ric Suter</a>, <a href="/search/cs?searchtype=author&query=Ward%2C+L">Logan Ward</a>, <a href="/search/cs?searchtype=author&query=Wilkinson%2C+S">Sean Wilkinson</a>, <a href="/search/cs?searchtype=author&query=Amaris%2C+M">Marcos Amaris</a>, <a href="/search/cs?searchtype=author&query=Babuji%2C+Y">Yadu Babuji</a>, <a href="/search/cs?searchtype=author&query=Bader%2C+J">Jonathan Bader</a>, <a href="/search/cs?searchtype=author&query=Balin%2C+R">Riccardo Balin</a>, <a href="/search/cs?searchtype=author&query=Balouek%2C+D">Daniel Balouek</a>, <a href="/search/cs?searchtype=author&query=Beecroft%2C+S">Sarah Beecroft</a>, <a href="/search/cs?searchtype=author&query=Belhajjame%2C+K">Khalid Belhajjame</a>, <a href="/search/cs?searchtype=author&query=Bhattarai%2C+R">Rajat Bhattarai</a> , et al. (86 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="2410.14943v1-abstract-short" style="display: inline;"> The Workflows Community Summit gathered 111 participants from 18 countries to discuss emerging trends and challenges in scientific workflows, focusing on six key areas: time-sensitive workflows, AI-HPC convergence, multi-facility workflows, heterogeneous HPC environments, user experience, and FAIR computational workflows. The integration of AI and exascale computing has revolutionized scientific w… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.14943v1-abstract-full').style.display = 'inline'; document.getElementById('2410.14943v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2410.14943v1-abstract-full" style="display: none;"> The Workflows Community Summit gathered 111 participants from 18 countries to discuss emerging trends and challenges in scientific workflows, focusing on six key areas: time-sensitive workflows, AI-HPC convergence, multi-facility workflows, heterogeneous HPC environments, user experience, and FAIR computational workflows. The integration of AI and exascale computing has revolutionized scientific workflows, enabling higher-fidelity models and complex, time-sensitive processes, while introducing challenges in managing heterogeneous environments and multi-facility data dependencies. The rise of large language models is driving computational demands to zettaflop scales, necessitating modular, adaptable systems and cloud-service models to optimize resource utilization and ensure reproducibility. Multi-facility workflows present challenges in data movement, curation, and overcoming institutional silos, while diverse hardware architectures require integrating workflow considerations into early system design and developing standardized resource management tools. The summit emphasized improving user experience in workflow systems and ensuring FAIR workflows to enhance collaboration and accelerate scientific discovery. Key recommendations include developing standardized metrics for time-sensitive workflows, creating frameworks for cloud-HPC integration, implementing distributed-by-design workflow modeling, establishing multi-facility authentication protocols, and accelerating AI integration in HPC workflow management. The summit also called for comprehensive workflow benchmarks, workflow-specific UX principles, and a FAIR workflow maturity model, highlighting the need for continued collaboration in addressing the complex challenges posed by the convergence of AI, HPC, and multi-facility research environments. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.14943v1-abstract-full').style.display = 'none'; document.getElementById('2410.14943v1-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, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Report number:</span> ORNL/TM-2024/3573 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2410.12927">arXiv:2410.12927</a> <span> [<a href="https://arxiv.org/pdf/2410.12927">pdf</a>, <a href="https://arxiv.org/format/2410.12927">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"> SoK: On Finding Common Ground in Loss Landscapes Using Deep Model Merging Techniques </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Khan%2C+A">Arham Khan</a>, <a href="/search/cs?searchtype=author&query=Nief%2C+T">Todd Nief</a>, <a href="/search/cs?searchtype=author&query=Hudson%2C+N">Nathaniel Hudson</a>, <a href="/search/cs?searchtype=author&query=Sakarvadia%2C+M">Mansi Sakarvadia</a>, <a href="/search/cs?searchtype=author&query=Grzenda%2C+D">Daniel Grzenda</a>, <a href="/search/cs?searchtype=author&query=Ajith%2C+A">Aswathy Ajith</a>, <a href="/search/cs?searchtype=author&query=Pettyjohn%2C+J">Jordan Pettyjohn</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</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="2410.12927v1-abstract-short" style="display: inline;"> Understanding neural networks is crucial to creating reliable and trustworthy deep learning models. Most contemporary research in interpretability analyzes just one model at a time via causal intervention or activation analysis. Yet despite successes, these methods leave significant gaps in our understanding of the training behaviors of neural networks, how their inner representations emerge, and… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.12927v1-abstract-full').style.display = 'inline'; document.getElementById('2410.12927v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2410.12927v1-abstract-full" style="display: none;"> Understanding neural networks is crucial to creating reliable and trustworthy deep learning models. Most contemporary research in interpretability analyzes just one model at a time via causal intervention or activation analysis. Yet despite successes, these methods leave significant gaps in our understanding of the training behaviors of neural networks, how their inner representations emerge, and how we can predictably associate model components with task-specific behaviors. Seeking new insights from work in related fields, here we survey literature in the field of model merging, a field that aims to combine the abilities of various neural networks by merging their parameters and identifying task-specific model components in the process. We analyze the model merging literature through the lens of loss landscape geometry, an approach that enables us to connect observations from empirical studies on interpretability, security, model merging, and loss landscape analysis to phenomena that govern neural network training and the emergence of their inner representations. To systematize knowledge in this area, we present a novel taxonomy of model merging techniques organized by their core algorithmic principles. Additionally, we distill repeated empirical observations from the literature in these fields into characterizations of four major aspects of loss landscape geometry: mode convexity, determinism, directedness, and connectivity. We argue that by improving our understanding of the principles underlying model merging and loss landscape geometry, this work contributes to the goal of ensuring secure and trustworthy machine learning in practice. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.12927v1-abstract-full').style.display = 'none'; document.getElementById('2410.12927v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 16 October, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2410.12092">arXiv:2410.12092</a> <span> [<a href="https://arxiv.org/pdf/2410.12092">pdf</a>, <a href="https://arxiv.org/format/2410.12092">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Distributed, Parallel, and Cluster Computing">cs.DC</span> </div> </div> <p class="title is-5 mathjax"> Accelerating Python Applications with Dask and ProxyStore </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pauloski%2C+J+G">J. Gregory Pauloski</a>, <a href="/search/cs?searchtype=author&query=Rydzy%2C+K">Klaudiusz Rydzy</a>, <a href="/search/cs?searchtype=author&query=Hayot-Sasson%2C+V">Valerie Hayot-Sasson</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</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="2410.12092v2-abstract-short" style="display: inline;"> Applications are increasingly written as dynamic workflows underpinned by an execution framework that manages asynchronous computations across distributed hardware. However, execution frameworks typically offer one-size-fits-all solutions for data flow management, which can restrict performance and scalability. ProxyStore, a middleware layer that optimizes data flow via an advanced pass-by-referen… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.12092v2-abstract-full').style.display = 'inline'; document.getElementById('2410.12092v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2410.12092v2-abstract-full" style="display: none;"> Applications are increasingly written as dynamic workflows underpinned by an execution framework that manages asynchronous computations across distributed hardware. However, execution frameworks typically offer one-size-fits-all solutions for data flow management, which can restrict performance and scalability. ProxyStore, a middleware layer that optimizes data flow via an advanced pass-by-reference paradigm, has shown to be an effective mechanism for addressing these limitations. Here, we investigate integrating ProxyStore with Dask Distributed, one of the most popular libraries for distributed computing in Python, with the goal of supporting scalable and portable scientific workflows. Dask provides an easy-to-use and flexible framework, but is less optimized for scaling certain data-intensive workflows. We investigate these limitations and detail the technical contributions necessary to develop a robust solution for distributed applications and demonstrate improved performance on synthetic benchmarks and real applications. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.12092v2-abstract-full').style.display = 'none'; document.getElementById('2410.12092v2-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 October, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 15 October, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">To be presented as a demo at the SC24 Workshop on High Performance Python for Science at Scale (HPPSS)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2410.02159">arXiv:2410.02159</a> <span> [<a href="https://arxiv.org/pdf/2410.02159">pdf</a>, <a href="https://arxiv.org/format/2410.02159">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="Computation and Language">cs.CL</span> </div> </div> <p class="title is-5 mathjax"> Mitigating Memorization In Language Models </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Sakarvadia%2C+M">Mansi Sakarvadia</a>, <a href="/search/cs?searchtype=author&query=Ajith%2C+A">Aswathy Ajith</a>, <a href="/search/cs?searchtype=author&query=Khan%2C+A">Arham Khan</a>, <a href="/search/cs?searchtype=author&query=Hudson%2C+N">Nathaniel Hudson</a>, <a href="/search/cs?searchtype=author&query=Geniesse%2C+C">Caleb Geniesse</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+Y">Yaoqing Yang</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=Mahoney%2C+M+W">Michael W. Mahoney</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="2410.02159v1-abstract-short" style="display: inline;"> Language models (LMs) can "memorize" information, i.e., encode training data in their weights in such a way that inference-time queries can lead to verbatim regurgitation of that data. This ability to extract training data can be problematic, for example, when data are private or sensitive. In this work, we investigate methods to mitigate memorization: three regularizer-based, three finetuning-bas… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.02159v1-abstract-full').style.display = 'inline'; document.getElementById('2410.02159v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2410.02159v1-abstract-full" style="display: none;"> Language models (LMs) can "memorize" information, i.e., encode training data in their weights in such a way that inference-time queries can lead to verbatim regurgitation of that data. This ability to extract training data can be problematic, for example, when data are private or sensitive. In this work, we investigate methods to mitigate memorization: three regularizer-based, three finetuning-based, and eleven machine unlearning-based methods, with five of the latter being new methods that we introduce. We also introduce TinyMem, a suite of small, computationally-efficient LMs for the rapid development and evaluation of memorization-mitigation methods. We demonstrate that the mitigation methods that we develop using TinyMem can successfully be applied to production-grade LMs, and we determine via experiment that: regularizer-based mitigation methods are slow and ineffective at curbing memorization; fine-tuning-based methods are effective at curbing memorization, but overly expensive, especially for retaining higher accuracies; and unlearning-based methods are faster and more effective, allowing for the precise localization and removal of memorized information from LM weights prior to inference. We show, in particular, that our proposed unlearning method BalancedSubnet outperforms other mitigation methods at removing memorized information while preserving performance on target tasks. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.02159v1-abstract-full').style.display = 'none'; document.getElementById('2410.02159v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 2 October, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2410.00709">arXiv:2410.00709</a> <span> [<a href="https://arxiv.org/pdf/2410.00709">pdf</a>, <a href="https://arxiv.org/format/2410.00709">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Quantitative Methods">q-bio.QM</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"> Binding Affinity Prediction: From Conventional to Machine Learning-Based Approaches </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Liu%2C+X">Xuefeng Liu</a>, <a href="/search/cs?searchtype=author&query=Jiang%2C+S">Songhao Jiang</a>, <a href="/search/cs?searchtype=author&query=Duan%2C+X">Xiaotian Duan</a>, <a href="/search/cs?searchtype=author&query=Vasan%2C+A">Archit Vasan</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+C">Chong Liu</a>, <a href="/search/cs?searchtype=author&query=Tien%2C+C">Chih-chan Tien</a>, <a href="/search/cs?searchtype=author&query=Ma%2C+H">Heng Ma</a>, <a href="/search/cs?searchtype=author&query=Brettin%2C+T">Thomas Brettin</a>, <a href="/search/cs?searchtype=author&query=Xia%2C+F">Fangfang Xia</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I+T">Ian T. Foster</a>, <a href="/search/cs?searchtype=author&query=Stevens%2C+R+L">Rick L. Stevens</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="2410.00709v1-abstract-short" style="display: inline;"> Protein-ligand binding is the process by which a small molecule (drug or inhibitor) attaches to a target protein. The binding affinity, which refers to the strength of this interaction, is central to many important problems in bioinformatics such as drug design. An extensive amount of work has been devoted to predicting binding affinity over the past decades due to its significance. In this paper,… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.00709v1-abstract-full').style.display = 'inline'; document.getElementById('2410.00709v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2410.00709v1-abstract-full" style="display: none;"> Protein-ligand binding is the process by which a small molecule (drug or inhibitor) attaches to a target protein. The binding affinity, which refers to the strength of this interaction, is central to many important problems in bioinformatics such as drug design. An extensive amount of work has been devoted to predicting binding affinity over the past decades due to its significance. In this paper, we review all significant recent works, focusing on the methods, features, and benchmark datasets. We have observed a rising trend in the use of traditional machine learning and deep learning models for predicting binding affinity, accompanied by an increasing amount of data on proteins and small drug-like molecules. While prediction results are constantly improving, we also identify several open questions and potential directions that remain unexplored in the field. This paper could serve as an excellent starting point for machine learning researchers who wish to engage in the study of binding affinity, or for anyone with general interests in machine learning, drug discovery, and bioinformatics. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.00709v1-abstract-full').style.display = 'none'; document.getElementById('2410.00709v1-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 September, 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/2409.16495">arXiv:2409.16495</a> <span> [<a href="https://arxiv.org/pdf/2409.16495">pdf</a>, <a href="https://arxiv.org/format/2409.16495">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="Distributed, Parallel, and Cluster Computing">cs.DC</span> </div> </div> <p class="title is-5 mathjax"> Flight: A FaaS-Based Framework for Complex and Hierarchical Federated Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Hudson%2C+N">Nathaniel Hudson</a>, <a href="/search/cs?searchtype=author&query=Hayot-Sasson%2C+V">Valerie Hayot-Sasson</a>, <a href="/search/cs?searchtype=author&query=Babuji%2C+Y">Yadu Babuji</a>, <a href="/search/cs?searchtype=author&query=Baughman%2C+M">Matt Baughman</a>, <a href="/search/cs?searchtype=author&query=Pauloski%2C+J+G">J. Gregory Pauloski</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+R">Ryan Chard</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</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="2409.16495v1-abstract-short" style="display: inline;"> Federated Learning (FL) is a decentralized machine learning paradigm where models are trained on distributed devices and are aggregated at a central server. Existing FL frameworks assume simple two-tier network topologies where end devices are directly connected to the aggregation server. While this is a practical mental model, it does not exploit the inherent topology of real-world distributed sy… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2409.16495v1-abstract-full').style.display = 'inline'; document.getElementById('2409.16495v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2409.16495v1-abstract-full" style="display: none;"> Federated Learning (FL) is a decentralized machine learning paradigm where models are trained on distributed devices and are aggregated at a central server. Existing FL frameworks assume simple two-tier network topologies where end devices are directly connected to the aggregation server. While this is a practical mental model, it does not exploit the inherent topology of real-world distributed systems like the Internet-of-Things. We present Flight, a novel FL framework that supports complex hierarchical multi-tier topologies, asynchronous aggregation, and decouples the control plane from the data plane. We compare the performance of Flight against Flower, a state-of-the-art FL framework. Our results show that Flight scales beyond Flower, supporting up to 2048 simultaneous devices, and reduces FL makespan across several models. Finally, we show that Flight's hierarchical FL model can reduce communication overheads by more than 60%. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2409.16495v1-abstract-full').style.display = 'none'; document.getElementById('2409.16495v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 24 September, 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/2408.14627">arXiv:2408.14627</a> <span> [<a href="https://arxiv.org/pdf/2408.14627">pdf</a>, <a href="https://arxiv.org/ps/2408.14627">ps</a>, <a href="https://arxiv.org/format/2408.14627">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Databases">cs.DB</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computational Engineering, Finance, and Science">cs.CE</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computers and Society">cs.CY</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Emerging Technologies">cs.ET</span> </div> </div> <p class="title is-5 mathjax"> Sustainable Data Democratization: A Multifaceted Investment for an Equitable Future </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Taufer%2C+M">Michela Taufer</a>, <a href="/search/cs?searchtype=author&query=Pascucci%2C+V">Valerio Pascucci</a>, <a href="/search/cs?searchtype=author&query=Kirkpatric%2C+C+R">Christine R. Kirkpatric</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I+T">Ian T. Foster</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="2408.14627v1-abstract-short" style="display: inline;"> The urgent need for data democratization in scientific research was the focal point of a panel discussion at SC23 in Denver, Colorado, from November 12 to 17, 2023. This article summarizes the outcomes of that discussion and subsequent conversations. We advocate for strategic investments in financial, human, and technological resources for sustainable data democratization. Emphasizing that data is… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2408.14627v1-abstract-full').style.display = 'inline'; document.getElementById('2408.14627v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2408.14627v1-abstract-full" style="display: none;"> The urgent need for data democratization in scientific research was the focal point of a panel discussion at SC23 in Denver, Colorado, from November 12 to 17, 2023. This article summarizes the outcomes of that discussion and subsequent conversations. We advocate for strategic investments in financial, human, and technological resources for sustainable data democratization. Emphasizing that data is central to scientific discovery and AI deployment, we highlight barriers such as limited access, inadequate financial incentives for cross-domain collaboration, and a shortage of workforce development initiatives. Our recommendations aim to guide decision-makers in fostering an inclusive research community, breaking down research silos, and developing a skilled workforce to advance scientific discovery. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2408.14627v1-abstract-full').style.display = 'none'; document.getElementById('2408.14627v1-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> 26 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">5 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/2408.14434">arXiv:2408.14434</a> <span> [<a href="https://arxiv.org/pdf/2408.14434">pdf</a>, <a href="https://arxiv.org/format/2408.14434">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Distributed, Parallel, and Cluster Computing">cs.DC</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"> Employing Artificial Intelligence to Steer Exascale Workflows with Colmena </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Ward%2C+L">Logan Ward</a>, <a href="/search/cs?searchtype=author&query=Pauloski%2C+J+G">J. Gregory Pauloski</a>, <a href="/search/cs?searchtype=author&query=Hayot-Sasson%2C+V">Valerie Hayot-Sasson</a>, <a href="/search/cs?searchtype=author&query=Babuji%2C+Y">Yadu Babuji</a>, <a href="/search/cs?searchtype=author&query=Brace%2C+A">Alexander Brace</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+R">Ryan Chard</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Thakur%2C+R">Rajeev Thakur</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</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="2408.14434v1-abstract-short" style="display: inline;"> Computational workflows are a common class of application on supercomputers, yet the loosely coupled and heterogeneous nature of workflows often fails to take full advantage of their capabilities. We created Colmena to leverage the massive parallelism of a supercomputer by using Artificial Intelligence (AI) to learn from and adapt a workflow as it executes. Colmena allows scientists to define how… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2408.14434v1-abstract-full').style.display = 'inline'; document.getElementById('2408.14434v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2408.14434v1-abstract-full" style="display: none;"> Computational workflows are a common class of application on supercomputers, yet the loosely coupled and heterogeneous nature of workflows often fails to take full advantage of their capabilities. We created Colmena to leverage the massive parallelism of a supercomputer by using Artificial Intelligence (AI) to learn from and adapt a workflow as it executes. Colmena allows scientists to define how their application should respond to events (e.g., task completion) as a series of cooperative agents. In this paper, we describe the design of Colmena, the challenges we overcame while deploying applications on exascale systems, and the science workflows we have enhanced through interweaving AI. The scaling challenges we discuss include developing steering strategies that maximize node utilization, introducing data fabrics that reduce communication overhead of data-intensive tasks, and implementing workflow tasks that cache costly operations between invocations. These innovations coupled with a variety of application patterns accessible through our agent-based steering model have enabled science advances in chemistry, biophysics, and materials science using different types of AI. Our vision is that Colmena will spur creative solutions that harness AI across many domains of scientific computing. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2408.14434v1-abstract-full').style.display = 'none'; document.getElementById('2408.14434v1-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> 26 August, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> August 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2408.07236">arXiv:2408.07236</a> <span> [<a href="https://arxiv.org/pdf/2408.07236">pdf</a>, <a href="https://arxiv.org/format/2408.07236">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Distributed, Parallel, and Cluster Computing">cs.DC</span> </div> </div> <p class="title is-5 mathjax"> TaPS: A Performance Evaluation Suite for Task-based Execution Frameworks </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pauloski%2C+J+G">J. Gregory Pauloski</a>, <a href="/search/cs?searchtype=author&query=Hayot-Sasson%2C+V">Valerie Hayot-Sasson</a>, <a href="/search/cs?searchtype=author&query=Gonthier%2C+M">Maxime Gonthier</a>, <a href="/search/cs?searchtype=author&query=Hudson%2C+N">Nathaniel Hudson</a>, <a href="/search/cs?searchtype=author&query=Pan%2C+H">Haochen Pan</a>, <a href="/search/cs?searchtype=author&query=Zhou%2C+S">Sicheng Zhou</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</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="2408.07236v1-abstract-short" style="display: inline;"> Task-based execution frameworks, such as parallel programming libraries, computational workflow systems, and function-as-a-service platforms, enable the composition of distinct tasks into a single, unified application designed to achieve a computational goal. Task-based execution frameworks abstract the parallel execution of an application's tasks on arbitrary hardware. Research into these task ex… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2408.07236v1-abstract-full').style.display = 'inline'; document.getElementById('2408.07236v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2408.07236v1-abstract-full" style="display: none;"> Task-based execution frameworks, such as parallel programming libraries, computational workflow systems, and function-as-a-service platforms, enable the composition of distinct tasks into a single, unified application designed to achieve a computational goal. Task-based execution frameworks abstract the parallel execution of an application's tasks on arbitrary hardware. Research into these task executors has accelerated as computational sciences increasingly need to take advantage of parallel compute and/or heterogeneous hardware. However, the lack of evaluation standards makes it challenging to compare and contrast novel systems against existing implementations. Here, we introduce TaPS, the Task Performance Suite, to support continued research in parallel task executor frameworks. TaPS provides (1) a unified, modular interface for writing and evaluating applications using arbitrary execution frameworks and data management systems and (2) an initial set of reference synthetic and real-world science applications. We discuss how the design of TaPS supports the reliable evaluation of frameworks and demonstrate TaPS through a survey of benchmarks using the provided reference applications. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2408.07236v1-abstract-full').style.display = 'none'; document.getElementById('2408.07236v1-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 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">To appear in the Proceedings of 20th IEEE International Conference on e-Science</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2407.11432">arXiv:2407.11432</a> <span> [<a href="https://arxiv.org/pdf/2407.11432">pdf</a>, <a href="https://arxiv.org/format/2407.11432">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Distributed, Parallel, and Cluster Computing">cs.DC</span> </div> </div> <p class="title is-5 mathjax"> Octopus: Experiences with a Hybrid Event-Driven Architecture for Distributed Scientific Computing </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pan%2C+H">Haochen Pan</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+R">Ryan Chard</a>, <a href="/search/cs?searchtype=author&query=Zhou%2C+S">Sicheng Zhou</a>, <a href="/search/cs?searchtype=author&query=Kamatar%2C+A">Alok Kamatar</a>, <a href="/search/cs?searchtype=author&query=Vescovi%2C+R">Rafael Vescovi</a>, <a href="/search/cs?searchtype=author&query=Hayot-Sasson%2C+V">Val茅rie Hayot-Sasson</a>, <a href="/search/cs?searchtype=author&query=Bauer%2C+A">Andr茅 Bauer</a>, <a href="/search/cs?searchtype=author&query=Gonthier%2C+M">Maxime Gonthier</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2407.11432v2-abstract-short" style="display: inline;"> Scientific research increasingly relies on distributed computational resources, storage systems, networks, and instruments, ranging from HPC and cloud systems to edge devices. Event-driven architecture (EDA) benefits applications targeting distributed research infrastructures by enabling the organization, communication, processing, reliability, and security of events generated from many sources. T… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2407.11432v2-abstract-full').style.display = 'inline'; document.getElementById('2407.11432v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2407.11432v2-abstract-full" style="display: none;"> Scientific research increasingly relies on distributed computational resources, storage systems, networks, and instruments, ranging from HPC and cloud systems to edge devices. Event-driven architecture (EDA) benefits applications targeting distributed research infrastructures by enabling the organization, communication, processing, reliability, and security of events generated from many sources. To support the development of scientific EDA, we introduce Octopus, a hybrid, cloud-to-edge event fabric designed to link many local event producers and consumers with cloud-hosted brokers. Octopus can be scaled to meet demand, permits the deployment of highly available Triggers for automatic event processing, and enforces fine-grained access control. We identify requirements in self-driving laboratories, scientific data automation, online task scheduling, epidemic modeling, and dynamic workflow management use cases, and present results demonstrating Octopus' ability to meet those requirements. Octopus supports producing and consuming events at a rate of over 4.2 M and 9.6 M events per second, respectively, from distributed clients. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2407.11432v2-abstract-full').style.display = 'none'; document.getElementById('2407.11432v2-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> 28 September, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 16 July, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">12 pages and 8 figures. Camera-ready version for FTXS'24 (https://sites.google.com/view/ftxs2024)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2407.09434">arXiv:2407.09434</a> <span> [<a href="https://arxiv.org/pdf/2407.09434">pdf</a>, <a href="https://arxiv.org/format/2407.09434">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="Computational Engineering, Finance, and Science">cs.CE</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Systems and Control">eess.SY</span> </div> </div> <p class="title is-5 mathjax"> Foundation Models for the Electric Power Grid </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Hamann%2C+H+F">Hendrik F. Hamann</a>, <a href="/search/cs?searchtype=author&query=Brunschwiler%2C+T">Thomas Brunschwiler</a>, <a href="/search/cs?searchtype=author&query=Gjorgiev%2C+B">Blazhe Gjorgiev</a>, <a href="/search/cs?searchtype=author&query=Martins%2C+L+S+A">Leonardo S. A. Martins</a>, <a href="/search/cs?searchtype=author&query=Puech%2C+A">Alban Puech</a>, <a href="/search/cs?searchtype=author&query=Varbella%2C+A">Anna Varbella</a>, <a href="/search/cs?searchtype=author&query=Weiss%2C+J">Jonas Weiss</a>, <a href="/search/cs?searchtype=author&query=Bernabe-Moreno%2C+J">Juan Bernabe-Moreno</a>, <a href="/search/cs?searchtype=author&query=Mass%C3%A9%2C+A+B">Alexandre Blondin Mass茅</a>, <a href="/search/cs?searchtype=author&query=Choi%2C+S">Seong Choi</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=Hodge%2C+B">Bri-Mathias Hodge</a>, <a href="/search/cs?searchtype=author&query=Jain%2C+R">Rishabh Jain</a>, <a href="/search/cs?searchtype=author&query=Kim%2C+K">Kibaek Kim</a>, <a href="/search/cs?searchtype=author&query=Mai%2C+V">Vincent Mai</a>, <a href="/search/cs?searchtype=author&query=Mirall%C3%A8s%2C+F">Fran莽ois Mirall猫s</a>, <a href="/search/cs?searchtype=author&query=De+Montigny%2C+M">Martin De Montigny</a>, <a href="/search/cs?searchtype=author&query=Ramos-Lea%C3%B1os%2C+O">Octavio Ramos-Lea帽os</a>, <a href="/search/cs?searchtype=author&query=Supr%C3%AAme%2C+H">Hussein Supr锚me</a>, <a href="/search/cs?searchtype=author&query=Xie%2C+L">Le Xie</a>, <a href="/search/cs?searchtype=author&query=Youssef%2C+E+S">El-Nasser S. Youssef</a>, <a href="/search/cs?searchtype=author&query=Zinflou%2C+A">Arnaud Zinflou</a>, <a href="/search/cs?searchtype=author&query=Belyi%2C+A+J">Alexander J. Belyi</a>, <a href="/search/cs?searchtype=author&query=Bessa%2C+R+J">Ricardo J. Bessa</a>, <a href="/search/cs?searchtype=author&query=Bhattarai%2C+B+P">Bishnu Prasad Bhattarai</a> , et al. (2 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="2407.09434v2-abstract-short" style="display: inline;"> Foundation models (FMs) currently dominate news headlines. They employ advanced deep learning architectures to extract structural information autonomously from vast datasets through self-supervision. The resulting rich representations of complex systems and dynamics can be applied to many downstream applications. Therefore, FMs can find uses in electric power grids, challenged by the energy transi… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2407.09434v2-abstract-full').style.display = 'inline'; document.getElementById('2407.09434v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2407.09434v2-abstract-full" style="display: none;"> Foundation models (FMs) currently dominate news headlines. They employ advanced deep learning architectures to extract structural information autonomously from vast datasets through self-supervision. The resulting rich representations of complex systems and dynamics can be applied to many downstream applications. Therefore, FMs can find uses in electric power grids, challenged by the energy transition and climate change. In this paper, we call for the development of, and state why we believe in, the potential of FMs for electric grids. We highlight their strengths and weaknesses amidst the challenges of a changing grid. We argue that an FM learning from diverse grid data and topologies could unlock transformative capabilities, pioneering a new approach in leveraging AI to redefine how we manage complexity and uncertainty in the electric grid. Finally, we discuss a power grid FM concept, namely GridFM, based on graph neural networks and show how different downstream tasks benefit. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2407.09434v2-abstract-full').style.display = 'none'; document.getElementById('2407.09434v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 12 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 12 July, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Major equal contributors: H.F.H., T.B., B.G., L.S.A.M., A.P., A.V., J.W.; Significant equal contributors: J.B., A.B.M., S.C., I.F., B.H., R.J., K.K., V.M., F.M., M.D.M., O.R., H.S., L.X., E.S.Y., A.Z.; Other equal contributors: A.J.B., R.J.B., B.P.B., J.S., S.S; Lead contact: H.F.H</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2407.01764">arXiv:2407.01764</a> <span> [<a href="https://arxiv.org/pdf/2407.01764">pdf</a>, <a href="https://arxiv.org/format/2407.01764">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Distributed, Parallel, and Cluster Computing">cs.DC</span> </div> </div> <p class="title is-5 mathjax"> Object Proxy Patterns for Accelerating Distributed Applications </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pauloski%2C+J+G">J. Gregory Pauloski</a>, <a href="/search/cs?searchtype=author&query=Hayot-Sasson%2C+V">Valerie Hayot-Sasson</a>, <a href="/search/cs?searchtype=author&query=Ward%2C+L">Logan Ward</a>, <a href="/search/cs?searchtype=author&query=Brace%2C+A">Alexander Brace</a>, <a href="/search/cs?searchtype=author&query=Bauer%2C+A">Andr茅 Bauer</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2407.01764v1-abstract-short" style="display: inline;"> Workflow and serverless frameworks have empowered new approaches to distributed application design by abstracting compute resources. However, their typically limited or one-size-fits-all support for advanced data flow patterns leaves optimization to the application programmer -- optimization that becomes more difficult as data become larger. The transparent object proxy, which provides wide-area r… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2407.01764v1-abstract-full').style.display = 'inline'; document.getElementById('2407.01764v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2407.01764v1-abstract-full" style="display: none;"> Workflow and serverless frameworks have empowered new approaches to distributed application design by abstracting compute resources. However, their typically limited or one-size-fits-all support for advanced data flow patterns leaves optimization to the application programmer -- optimization that becomes more difficult as data become larger. The transparent object proxy, which provides wide-area references that can resolve to data regardless of location, has been demonstrated as an effective low-level building block in such situations. Here we propose three high-level proxy-based programming patterns -- distributed futures, streaming, and ownership -- that make the power of the proxy pattern usable for more complex and dynamic distributed program structures. We motivate these patterns via careful review of application requirements and describe implementations of each pattern. We evaluate our implementations through a suite of benchmarks and by applying them in three substantial scientific applications, in which we demonstrate substantial improvements in runtime, throughput, and memory usage. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2407.01764v1-abstract-full').style.display = 'none'; document.getElementById('2407.01764v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 1 July, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2406.17710">arXiv:2406.17710</a> <span> [<a href="https://arxiv.org/pdf/2406.17710">pdf</a>, <a href="https://arxiv.org/format/2406.17710">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Distributed, Parallel, and Cluster Computing">cs.DC</span> </div> </div> <p class="title is-5 mathjax"> GreenFaaS: Maximizing Energy Efficiency of HPC Workloads with FaaS </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Kamatar%2C+A">Alok Kamatar</a>, <a href="/search/cs?searchtype=author&query=Hayot-Sasson%2C+V">Valerie Hayot-Sasson</a>, <a href="/search/cs?searchtype=author&query=Babuji%2C+Y">Yadu Babuji</a>, <a href="/search/cs?searchtype=author&query=Bauer%2C+A">Andre Bauer</a>, <a href="/search/cs?searchtype=author&query=Rattihalli%2C+G">Gourav Rattihalli</a>, <a href="/search/cs?searchtype=author&query=Hogade%2C+N">Ninad Hogade</a>, <a href="/search/cs?searchtype=author&query=Milojicic%2C+D">Dejan Milojicic</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2406.17710v1-abstract-short" style="display: inline;"> Application energy efficiency can be improved by executing each application component on the compute element that consumes the least energy while also satisfying time constraints. In principle, the function as a service (FaaS) paradigm should simplify such optimizations by abstracting away compute location, but existing FaaS systems do not provide for user transparency over application energy cons… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2406.17710v1-abstract-full').style.display = 'inline'; document.getElementById('2406.17710v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2406.17710v1-abstract-full" style="display: none;"> Application energy efficiency can be improved by executing each application component on the compute element that consumes the least energy while also satisfying time constraints. In principle, the function as a service (FaaS) paradigm should simplify such optimizations by abstracting away compute location, but existing FaaS systems do not provide for user transparency over application energy consumption or task placement. Here we present GreenFaaS, a novel open source framework that bridges this gap between energy-efficient applications and FaaS platforms. GreenFaaS can be deployed by end users or providers across systems to monitor energy use, provide task-specific feedback, and schedule tasks in an energy-aware manner. We demonstrate that intelligent placement of tasks can both reduce energy consumption and improve performance. For a synthetic workload, GreenFaaS reduces the energy-delay product by 45% compared to alternatives. Furthermore, running a molecular design application through GreenFaaS can reduce energy consumption by 21% and runtime by 63% by better matching tasks with machines. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2406.17710v1-abstract-full').style.display = 'none'; document.getElementById('2406.17710v1-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, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">11 pages, 10 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/2406.06348">arXiv:2406.06348</a> <span> [<a href="https://arxiv.org/pdf/2406.06348">pdf</a>, <a href="https://arxiv.org/format/2406.06348">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="Distributed, Parallel, and Cluster Computing">cs.DC</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Methodology">stat.ME</span> </div> </div> <p class="title is-5 mathjax"> Causal Discovery over High-Dimensional Structured Hypothesis Spaces with Causal Graph Partitioning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Shah%2C+A">Ashka Shah</a>, <a href="/search/cs?searchtype=author&query=DePavia%2C+A">Adela DePavia</a>, <a href="/search/cs?searchtype=author&query=Hudson%2C+N">Nathaniel Hudson</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=Stevens%2C+R">Rick Stevens</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2406.06348v2-abstract-short" style="display: inline;"> The aim in many sciences is to understand the mechanisms that underlie the observed distribution of variables, starting from a set of initial hypotheses. Causal discovery allows us to infer mechanisms as sets of cause and effect relationships in a generalized way -- without necessarily tailoring to a specific domain. Causal discovery algorithms search over a structured hypothesis space, defined by… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2406.06348v2-abstract-full').style.display = 'inline'; document.getElementById('2406.06348v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2406.06348v2-abstract-full" style="display: none;"> The aim in many sciences is to understand the mechanisms that underlie the observed distribution of variables, starting from a set of initial hypotheses. Causal discovery allows us to infer mechanisms as sets of cause and effect relationships in a generalized way -- without necessarily tailoring to a specific domain. Causal discovery algorithms search over a structured hypothesis space, defined by the set of directed acyclic graphs, to find the graph that best explains the data. For high-dimensional problems, however, this search becomes intractable and scalable algorithms for causal discovery are needed to bridge the gap. In this paper, we define a novel causal graph partition that allows for divide-and-conquer causal discovery with theoretical guarantees. We leverage the idea of a superstructure -- a set of learned or existing candidate hypotheses -- to partition the search space. We prove under certain assumptions that learning with a causal graph partition always yields the Markov Equivalence Class of the true causal graph. We show our algorithm achieves comparable accuracy and a faster time to solution for biologically-tuned synthetic networks and networks up to ${10^4}$ variables. This makes our method applicable to gene regulatory network inference and other domains with high-dimensional structured hypothesis spaces. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2406.06348v2-abstract-full').style.display = 'none'; document.getElementById('2406.06348v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 24 July, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 10 June, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2406.03285">arXiv:2406.03285</a> <span> [<a href="https://arxiv.org/pdf/2406.03285">pdf</a>, <a href="https://arxiv.org/format/2406.03285">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Distributed, Parallel, and Cluster Computing">cs.DC</span> </div> <div class="is-inline-block" style="margin-left: 0.5rem"> <div class="tags has-addons"> <span class="tag is-dark is-size-7">doi</span> <span class="tag is-light is-size-7"><a class="" href="https://doi.org/10.1109/CCGrid59990.2024.00036">10.1109/CCGrid59990.2024.00036 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Efficient Data-Parallel Continual Learning with Asynchronous Distributed Rehearsal Buffers </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Bouvier%2C+T">Thomas Bouvier</a>, <a href="/search/cs?searchtype=author&query=Nicolae%2C+B">Bogdan Nicolae</a>, <a href="/search/cs?searchtype=author&query=Chaugier%2C+H">Hugo Chaugier</a>, <a href="/search/cs?searchtype=author&query=Costan%2C+A">Alexandru Costan</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=Antoniu%2C+G">Gabriel Antoniu</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2406.03285v1-abstract-short" style="display: inline;"> Deep learning has emerged as a powerful method for extracting valuable information from large volumes of data. However, when new training data arrives continuously (i.e., is not fully available from the beginning), incremental training suffers from catastrophic forgetting (i.e., new patterns are reinforced at the expense of previously acquired knowledge). Training from scratch each time new traini… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2406.03285v1-abstract-full').style.display = 'inline'; document.getElementById('2406.03285v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2406.03285v1-abstract-full" style="display: none;"> Deep learning has emerged as a powerful method for extracting valuable information from large volumes of data. However, when new training data arrives continuously (i.e., is not fully available from the beginning), incremental training suffers from catastrophic forgetting (i.e., new patterns are reinforced at the expense of previously acquired knowledge). Training from scratch each time new training data becomes available would result in extremely long training times and massive data accumulation. Rehearsal-based continual learning has shown promise for addressing the catastrophic forgetting challenge, but research to date has not addressed performance and scalability. To fill this gap, we propose an approach based on a distributed rehearsal buffer that efficiently complements data-parallel training on multiple GPUs, allowing us to achieve short runtime and scalability while retaining high accuracy. It leverages a set of buffers (local to each GPU) and uses several asynchronous techniques for updating these local buffers in an embarrassingly parallel fashion, all while handling the communication overheads necessary to augment input mini-batches (groups of training samples fed to the model) using unbiased, global sampling. In this paper we explore the benefits of this approach for classification models. We run extensive experiments on up to 128 GPUs of the ThetaGPU supercomputer to compare our approach with baselines representative of training-from-scratch (the upper bound in terms of accuracy) and incremental training (the lower bound). Results show that rehearsal-based continual learning achieves a top-5 classification accuracy close to the upper bound, while simultaneously exhibiting a runtime close to the lower bound. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2406.03285v1-abstract-full').style.display = 'none'; document.getElementById('2406.03285v1-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 June, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">2024 IEEE 24th International Symposium on Cluster, Cloud and Internet Computing (CCGrid), May 2024, Philadelphia (PA), United States</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2405.15828">arXiv:2405.15828</a> <span> [<a href="https://arxiv.org/pdf/2405.15828">pdf</a>, <a href="https://arxiv.org/format/2405.15828">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Digital Libraries">cs.DL</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"> Oil & Water? Diffusion of AI Within and Across Scientific Fields </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Duede%2C+E">Eamon Duede</a>, <a href="/search/cs?searchtype=author&query=Dolan%2C+W">William Dolan</a>, <a href="/search/cs?searchtype=author&query=Bauer%2C+A">Andr茅 Bauer</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=Lakhani%2C+K">Karim Lakhani</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="2405.15828v1-abstract-short" style="display: inline;"> This study empirically investigates claims of the increasing ubiquity of artificial intelligence (AI) within roughly 80 million research publications across 20 diverse scientific fields, by examining the change in scholarly engagement with AI from 1985 through 2022. We observe exponential growth, with AI-engaged publications increasing approximately thirteenfold (13x) across all fields, suggesting… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2405.15828v1-abstract-full').style.display = 'inline'; document.getElementById('2405.15828v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2405.15828v1-abstract-full" style="display: none;"> This study empirically investigates claims of the increasing ubiquity of artificial intelligence (AI) within roughly 80 million research publications across 20 diverse scientific fields, by examining the change in scholarly engagement with AI from 1985 through 2022. We observe exponential growth, with AI-engaged publications increasing approximately thirteenfold (13x) across all fields, suggesting a dramatic shift from niche to mainstream. Moreover, we provide the first empirical examination of the distribution of AI-engaged publications across publication venues within individual fields, with results that reveal a broadening of AI engagement within disciplines. While this broadening engagement suggests a move toward greater disciplinary integration in every field, increased ubiquity is associated with a semantic tension between AI-engaged research and more traditional disciplinary research. Through an analysis of tens of millions of document embeddings, we observe a complex interplay between AI-engaged and non-AI-engaged research within and across fields, suggesting that increasing ubiquity is something of an oil-and-water phenomenon -- AI-engaged work is spreading out over fields, but not mixing well with non-AI-engaged work. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2405.15828v1-abstract-full').style.display = 'none'; document.getElementById('2405.15828v1-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 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/2405.09939">arXiv:2405.09939</a> <span> [<a href="https://arxiv.org/pdf/2405.09939">pdf</a>, <a href="https://arxiv.org/format/2405.09939">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computation and Language">cs.CL</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> SciQAG: A Framework for Auto-Generated Science Question Answering Dataset with Fine-grained Evaluation </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Wan%2C+Y">Yuwei Wan</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+Y">Yixuan Liu</a>, <a href="/search/cs?searchtype=author&query=Ajith%2C+A">Aswathy Ajith</a>, <a href="/search/cs?searchtype=author&query=Grazian%2C+C">Clara Grazian</a>, <a href="/search/cs?searchtype=author&query=Hoex%2C+B">Bram Hoex</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+W">Wenjie Zhang</a>, <a href="/search/cs?searchtype=author&query=Kit%2C+C">Chunyu Kit</a>, <a href="/search/cs?searchtype=author&query=Xie%2C+T">Tong Xie</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</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="2405.09939v2-abstract-short" style="display: inline;"> We introduce SciQAG, a novel framework for automatically generating high-quality science question-answer pairs from a large corpus of scientific literature based on large language models (LLMs). SciQAG consists of a QA generator and a QA evaluator, which work together to extract diverse and research-level questions and answers from scientific papers. Utilizing this framework, we construct a large-… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2405.09939v2-abstract-full').style.display = 'inline'; document.getElementById('2405.09939v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2405.09939v2-abstract-full" style="display: none;"> We introduce SciQAG, a novel framework for automatically generating high-quality science question-answer pairs from a large corpus of scientific literature based on large language models (LLMs). SciQAG consists of a QA generator and a QA evaluator, which work together to extract diverse and research-level questions and answers from scientific papers. Utilizing this framework, we construct a large-scale, high-quality, open-ended science QA dataset containing 188,042 QA pairs extracted from 22,743 scientific papers across 24 scientific domains. We also introduce SciQAG-24D, a new benchmark task designed to evaluate the science question-answering ability of LLMs. Extensive experiments demonstrate that fine-tuning LLMs on the SciQAG dataset significantly improves their performance on both open-ended question answering and scientific tasks. To foster research and collaboration, we make the datasets, models, and evaluation codes publicly available, contributing to the advancement of science question answering and developing more interpretable and reasoning-capable AI systems. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2405.09939v2-abstract-full').style.display = 'none'; document.getElementById('2405.09939v2-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 July, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 16 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/2404.19717">arXiv:2404.19717</a> <span> [<a href="https://arxiv.org/pdf/2404.19717">pdf</a>, <a href="https://arxiv.org/format/2404.19717">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Distributed, Parallel, and Cluster Computing">cs.DC</span> </div> </div> <p class="title is-5 mathjax"> Automated, Reliable, and Efficient Continental-Scale Replication of 7.3 Petabytes of Climate Simulation Data: A Case Study </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lacinski%2C+L">Lukasz Lacinski</a>, <a href="/search/cs?searchtype=author&query=Liming%2C+L">Lee Liming</a>, <a href="/search/cs?searchtype=author&query=Turoscy%2C+S">Steven Turoscy</a>, <a href="/search/cs?searchtype=author&query=Harr%2C+C">Cameron Harr</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Dart%2C+E">Eli Dart</a>, <a href="/search/cs?searchtype=author&query=Durack%2C+P">Paul Durack</a>, <a href="/search/cs?searchtype=author&query=Ames%2C+S">Sasha Ames</a>, <a href="/search/cs?searchtype=author&query=Hoffman%2C+F+M">Forrest M. Hoffman</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I+T">Ian T. Foster</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2404.19717v1-abstract-short" style="display: inline;"> We report on our experiences replicating 7.3 petabytes (PB) of Earth System Grid Federation (ESGF) climate simulation data from Lawrence Livermore National Laboratory (LLNL) in California to Argonne National Laboratory (ANL) in Illinois and Oak Ridge National Laboratory (ORNL) in Tennessee. This movement of some 29 million files, twice, undertaken in order to establish new ESGF nodes at ANL and OR… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2404.19717v1-abstract-full').style.display = 'inline'; document.getElementById('2404.19717v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2404.19717v1-abstract-full" style="display: none;"> We report on our experiences replicating 7.3 petabytes (PB) of Earth System Grid Federation (ESGF) climate simulation data from Lawrence Livermore National Laboratory (LLNL) in California to Argonne National Laboratory (ANL) in Illinois and Oak Ridge National Laboratory (ORNL) in Tennessee. This movement of some 29 million files, twice, undertaken in order to establish new ESGF nodes at ANL and ORNL, was performed largely automatically by a simple replication tool, a script that invoked Globus to transfer large bundles of files while tracking progress in a database. Under the covers, Globus organized transfers to make efficient use of the high-speed Energy Sciences network (ESnet) and the data transfer nodes deployed at participating sites, and also addressed security, integrity checking, and recovery from a variety of transient failures. This success demonstrates the considerable benefits that can accrue from the adoption of performant data replication infrastructure. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2404.19717v1-abstract-full').style.display = 'none'; document.getElementById('2404.19717v1-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 April, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2404.15668">arXiv:2404.15668</a> <span> [<a href="https://arxiv.org/pdf/2404.15668">pdf</a>, <a href="https://arxiv.org/format/2404.15668">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Distributed, Parallel, and Cluster Computing">cs.DC</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/10.1145/3629526.3645035">10.1145/3629526.3645035 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> MalleTrain: Deep Neural Network Training on Unfillable Supercomputer Nodes </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Ma%2C+X">Xiaolong Ma</a>, <a href="/search/cs?searchtype=author&query=Yan%2C+F">Feng Yan</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+L">Lei Yang</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=Papka%2C+M+E">Michael E. Papka</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+Z">Zhengchun Liu</a>, <a href="/search/cs?searchtype=author&query=Kettimuthu%2C+R">Rajkumar Kettimuthu</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2404.15668v1-abstract-short" style="display: inline;"> First-come first-serve scheduling can result in substantial (up to 10%) of transiently idle nodes on supercomputers. Recognizing that such unfilled nodes are well-suited for deep neural network (DNN) training, due to the flexible nature of DNN training tasks, Liu et al. proposed that the re-scaling DNN training tasks to fit gaps in schedules be formulated as a mixed-integer linear programming (MIL… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2404.15668v1-abstract-full').style.display = 'inline'; document.getElementById('2404.15668v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2404.15668v1-abstract-full" style="display: none;"> First-come first-serve scheduling can result in substantial (up to 10%) of transiently idle nodes on supercomputers. Recognizing that such unfilled nodes are well-suited for deep neural network (DNN) training, due to the flexible nature of DNN training tasks, Liu et al. proposed that the re-scaling DNN training tasks to fit gaps in schedules be formulated as a mixed-integer linear programming (MILP) problem, and demonstrated via simulation the potential benefits of the approach. Here, we introduce MalleTrain, a system that provides the first practical implementation of this approach and that furthermore generalizes it by allowing it use even for DNN training applications for which model information is unknown before runtime. Key to this latter innovation is the use of a lightweight online job profiling advisor (JPA) to collect critical scalability information for DNN jobs -- information that it then employs to optimize resource allocations dynamically, in real time. We describe the MalleTrain architecture and present the results of a detailed experimental evaluation on a supercomputer GPU cluster and several representative DNN training workloads, including neural architecture search and hyperparameter optimization. Our results not only confirm the practical feasibility of leveraging idle supercomputer nodes for DNN training but improve significantly on prior results, improving training throughput by up to 22.3\% without requiring users to provide job scalability information. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2404.15668v1-abstract-full').style.display = 'none'; document.getElementById('2404.15668v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 24 April, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2404.04225">arXiv:2404.04225</a> <span> [<a href="https://arxiv.org/pdf/2404.04225">pdf</a>, <a href="https://arxiv.org/format/2404.04225">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Chemical Physics">physics.chem-ph</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"> Twins in rotational spectroscopy: Does a rotational spectrum uniquely identify a molecule? </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Schwarting%2C+M">Marcus Schwarting</a>, <a href="/search/cs?searchtype=author&query=Seifert%2C+N+A">Nathan A. Seifert</a>, <a href="/search/cs?searchtype=author&query=Davis%2C+M+J">Michael J. Davis</a>, <a href="/search/cs?searchtype=author&query=Blaiszik%2C+B">Ben Blaiszik</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=Prozument%2C+K">Kirill Prozument</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2404.04225v1-abstract-short" style="display: inline;"> Rotational spectroscopy is the most accurate method for determining structures of molecules in the gas phase. It is often assumed that a rotational spectrum is a unique "fingerprint" of a molecule. The availability of large molecular databases and the development of artificial intelligence methods for spectroscopy makes the testing of this assumption timely. In this paper, we pose the determinatio… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2404.04225v1-abstract-full').style.display = 'inline'; document.getElementById('2404.04225v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2404.04225v1-abstract-full" style="display: none;"> Rotational spectroscopy is the most accurate method for determining structures of molecules in the gas phase. It is often assumed that a rotational spectrum is a unique "fingerprint" of a molecule. The availability of large molecular databases and the development of artificial intelligence methods for spectroscopy makes the testing of this assumption timely. In this paper, we pose the determination of molecular structures from rotational spectra as an inverse problem. Within this framework, we adopt a funnel-based approach to search for molecular twins, which are two or more molecules, which have similar rotational spectra but distinctly different molecular structures. We demonstrate that there are twins within standard levels of computational accuracy by generating rotational constants for many molecules from several large molecular databases, indicating the inverse problem is ill-posed. However, some twins can be distinguished by increasing the accuracy of the theoretical methods or by performing additional experiments. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2404.04225v1-abstract-full').style.display = 'none'; document.getElementById('2404.04225v1-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 April, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2404.02163">arXiv:2404.02163</a> <span> [<a href="https://arxiv.org/pdf/2404.02163">pdf</a>, <a href="https://arxiv.org/format/2404.02163">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Information Theory">cs.IT</span> </div> </div> <p class="title is-5 mathjax"> FastqZip: An Improved Reference-Based Genome Sequence Lossy Compression Framework </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Liu%2C+Y">Yuanjian Liu</a>, <a href="/search/cs?searchtype=author&query=Luo%2C+H">Huihao Luo</a>, <a href="/search/cs?searchtype=author&query=Han%2C+Z">Zhijun Han</a>, <a href="/search/cs?searchtype=author&query=Hu%2C+Y">Yao Hu</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+Y">Yehui Yang</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Di%2C+S">Sheng Di</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=Wu%2C+J">Jiesheng Wu</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2404.02163v1-abstract-short" style="display: inline;"> Storing and archiving data produced by next-generation sequencing (NGS) is a huge burden for research institutions. Reference-based compression algorithms are effective in dealing with these data. Our work focuses on compressing FASTQ format files with an improved reference-based compression algorithm to achieve a higher compression ratio than other state-of-the-art algorithms. We propose FastqZip… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2404.02163v1-abstract-full').style.display = 'inline'; document.getElementById('2404.02163v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2404.02163v1-abstract-full" style="display: none;"> Storing and archiving data produced by next-generation sequencing (NGS) is a huge burden for research institutions. Reference-based compression algorithms are effective in dealing with these data. Our work focuses on compressing FASTQ format files with an improved reference-based compression algorithm to achieve a higher compression ratio than other state-of-the-art algorithms. We propose FastqZip, which uses a new method mapping the sequence to reference for compression, allows reads-reordering and lossy quality scores, and the BSC or ZPAQ algorithm to perform final lossless compression for a higher compression ratio and relatively fast speed. Our method ensures the sequence can be losslessly reconstructed while allowing lossless or lossy compression for the quality scores. We reordered the reads to get a higher compression ratio. We evaluate our algorithms on five datasets and show that FastqZip can outperform the SOTA algorithm Genozip by around 10% in terms of compression ratio while having an acceptable slowdown. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2404.02163v1-abstract-full').style.display = 'none'; document.getElementById('2404.02163v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 22 February, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2403.19257">arXiv:2403.19257</a> <span> [<a href="https://arxiv.org/pdf/2403.19257">pdf</a>, <a href="https://arxiv.org/format/2403.19257">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Distributed, Parallel, and Cluster Computing">cs.DC</span> </div> <div class="is-inline-block" style="margin-left: 0.5rem"> <div class="tags has-addons"> <span class="tag is-dark is-size-7">doi</span> <span class="tag is-light is-size-7"><a class="" href="https://doi.org/10.1109/IPDPS57955.2024.00027">10.1109/IPDPS57955.2024.00027 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> UniFaaS: Programming across Distributed Cyberinfrastructure with Federated Function Serving </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Li%2C+Y">Yifei Li</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+R">Ryan Chard</a>, <a href="/search/cs?searchtype=author&query=Babuji%2C+Y">Yadu Babuji</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=Li%2C+Z">Zhuozhao Li</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2403.19257v1-abstract-short" style="display: inline;"> Modern scientific applications are increasingly decomposable into individual functions that may be deployed across distributed and diverse cyberinfrastructure such as supercomputers, clouds, and accelerators. Such applications call for new approaches to programming, distributed execution, and function-level management. We present UniFaaS, a parallel programming framework that relies on a federated… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2403.19257v1-abstract-full').style.display = 'inline'; document.getElementById('2403.19257v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2403.19257v1-abstract-full" style="display: none;"> Modern scientific applications are increasingly decomposable into individual functions that may be deployed across distributed and diverse cyberinfrastructure such as supercomputers, clouds, and accelerators. Such applications call for new approaches to programming, distributed execution, and function-level management. We present UniFaaS, a parallel programming framework that relies on a federated function-as-a-service (FaaS) model to enable composition of distributed, scalable, and high-performance scientific workflows, and to support fine-grained function-level management. UniFaaS provides a unified programming interface to compose dynamic task graphs with transparent wide-area data management. UniFaaS exploits an observe-predict-decide approach to efficiently map workflow tasks to target heterogeneous and dynamic resources. We propose a dynamic heterogeneity-aware scheduling algorithm that employs a delay mechanism and a re-scheduling mechanism to accommodate dynamic resource capacity. Our experiments show that UniFaaS can efficiently execute workflows across computing resources with minimal scheduling overhead. We show that UniFaaS can improve the performance of a real-world drug screening workflow by as much as 22.99% when employing an additional 19.48% of resources and a montage workflow by 54.41% when employing an additional 47.83% of resources across multiple distributed clusters, in contrast to using a single cluster <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2403.19257v1-abstract-full').style.display = 'none'; document.getElementById('2403.19257v1-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> 28 March, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">13 pages, 13 figures, IPDPS2024</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2403.06077">arXiv:2403.06077</a> <span> [<a href="https://arxiv.org/pdf/2403.06077">pdf</a>, <a href="https://arxiv.org/format/2403.06077">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Distributed, Parallel, and Cluster Computing">cs.DC</span> </div> </div> <p class="title is-5 mathjax"> Steering a Fleet: Adaptation for Large-Scale, Workflow-Based Experiments </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pruyne%2C+J">Jim Pruyne</a>, <a href="/search/cs?searchtype=author&query=Hayot-Sasson%2C+V">Valerie Hayot-Sasson</a>, <a href="/search/cs?searchtype=author&query=Zheng%2C+W">Weijian Zheng</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+R">Ryan Chard</a>, <a href="/search/cs?searchtype=author&query=Wozniak%2C+J+M">Justin M. Wozniak</a>, <a href="/search/cs?searchtype=author&query=Bicer%2C+T">Tekin Bicer</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I+T">Ian T. Foster</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2403.06077v1-abstract-short" style="display: inline;"> Experimental science is increasingly driven by instruments that produce vast volumes of data and thus a need to manage, compute, describe, and index this data. High performance and distributed computing provide the means of addressing the computing needs; however, in practice, the variety of actions required and the distributed set of resources involved, requires sophisticated "flows" defining the… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2403.06077v1-abstract-full').style.display = 'inline'; document.getElementById('2403.06077v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2403.06077v1-abstract-full" style="display: none;"> Experimental science is increasingly driven by instruments that produce vast volumes of data and thus a need to manage, compute, describe, and index this data. High performance and distributed computing provide the means of addressing the computing needs; however, in practice, the variety of actions required and the distributed set of resources involved, requires sophisticated "flows" defining the steps to be performed on data. As each scan or measurement is performed by an instrument, a new instance of the flow is initiated resulting in a "fleet" of concurrently running flows, with the overall goal to process all the data collected during a potentially long-running experiment. During the course of the experiment, each flow may need to adapt its execution due to changes in the environment, such as computational or storage resource availability, or based on the progress of the fleet as a whole such as completion or discovery of an intermediate result leading to a change in subsequent flow's behavior. We introduce a cloud-based decision engine, Braid, which flows consult during execution to query their run-time environment and coordinate with other flows within their fleet. Braid accepts streams of measurements taken from the run-time environment or from within flow runs which can then be statistically aggregated and compared to other streams to determine a strategy to guide flow execution. For example, queue lengths in execution environments can be used to direct a flow to run computations in one environment or another, or experiment progress as measured by individual flows can be aggregated to determine the progress and subsequent direction of the flows within a fleet. We describe Braid, its interface, implementation and performance characteristics. We further show through examples and experience modifying an existing scientific flow how Braid is used to make adaptable flows. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2403.06077v1-abstract-full').style.display = 'none'; document.getElementById('2403.06077v1-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 March, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2402.14129">arXiv:2402.14129</a> <span> [<a href="https://arxiv.org/pdf/2402.14129">pdf</a>, <a href="https://arxiv.org/format/2402.14129">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Information Retrieval">cs.IR</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computation and Language">cs.CL</span> </div> </div> <p class="title is-5 mathjax"> Combining Language and Graph Models for Semi-structured Information Extraction on the Web </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Hong%2C+Z">Zhi Hong</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</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="2402.14129v1-abstract-short" style="display: inline;"> Relation extraction is an efficient way of mining the extraordinary wealth of human knowledge on the Web. Existing methods rely on domain-specific training data or produce noisy outputs. We focus here on extracting targeted relations from semi-structured web pages given only a short description of the relation. We present GraphScholarBERT, an open-domain information extraction method based on a jo… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2402.14129v1-abstract-full').style.display = 'inline'; document.getElementById('2402.14129v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2402.14129v1-abstract-full" style="display: none;"> Relation extraction is an efficient way of mining the extraordinary wealth of human knowledge on the Web. Existing methods rely on domain-specific training data or produce noisy outputs. We focus here on extracting targeted relations from semi-structured web pages given only a short description of the relation. We present GraphScholarBERT, an open-domain information extraction method based on a joint graph and language model structure. GraphScholarBERT can generalize to previously unseen domains without additional data or training and produces only clean extraction results matched to the search keyword. Experiments show that GraphScholarBERT can improve extraction F1 scores by as much as 34.8\% compared to previous work in a zero-shot domain and zero-shot website setting. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2402.14129v1-abstract-full').style.display = 'none'; document.getElementById('2402.14129v1-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 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">7 pages, 2 figures</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2402.03480">arXiv:2402.03480</a> <span> [<a href="https://arxiv.org/pdf/2402.03480">pdf</a>, <a href="https://arxiv.org/format/2402.03480">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Distributed, Parallel, and Cluster Computing">cs.DC</span> </div> </div> <p class="title is-5 mathjax"> Trillion Parameter AI Serving Infrastructure for Scientific Discovery: A Survey and Vision </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Hudson%2C+N">Nathaniel Hudson</a>, <a href="/search/cs?searchtype=author&query=Pauloski%2C+J+G">J. Gregory Pauloski</a>, <a href="/search/cs?searchtype=author&query=Baughman%2C+M">Matt Baughman</a>, <a href="/search/cs?searchtype=author&query=Kamatar%2C+A">Alok Kamatar</a>, <a href="/search/cs?searchtype=author&query=Sakarvadia%2C+M">Mansi Sakarvadia</a>, <a href="/search/cs?searchtype=author&query=Ward%2C+L">Logan Ward</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+R">Ryan Chard</a>, <a href="/search/cs?searchtype=author&query=Bauer%2C+A">Andr茅 Bauer</a>, <a href="/search/cs?searchtype=author&query=Levental%2C+M">Maksim Levental</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+W">Wenyi Wang</a>, <a href="/search/cs?searchtype=author&query=Engler%2C+W">Will Engler</a>, <a href="/search/cs?searchtype=author&query=Skelly%2C+O+P">Owen Price Skelly</a>, <a href="/search/cs?searchtype=author&query=Blaiszik%2C+B">Ben Blaiszik</a>, <a href="/search/cs?searchtype=author&query=Stevens%2C+R">Rick Stevens</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</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="2402.03480v1-abstract-short" style="display: inline;"> Deep learning methods are transforming research, enabling new techniques, and ultimately leading to new discoveries. As the demand for more capable AI models continues to grow, we are now entering an era of Trillion Parameter Models (TPM), or models with more than a trillion parameters -- such as Huawei's PanGu-$危$. We describe a vision for the ecosystem of TPM users and providers that caters to t… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2402.03480v1-abstract-full').style.display = 'inline'; document.getElementById('2402.03480v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2402.03480v1-abstract-full" style="display: none;"> Deep learning methods are transforming research, enabling new techniques, and ultimately leading to new discoveries. As the demand for more capable AI models continues to grow, we are now entering an era of Trillion Parameter Models (TPM), or models with more than a trillion parameters -- such as Huawei's PanGu-$危$. We describe a vision for the ecosystem of TPM users and providers that caters to the specific needs of the scientific community. We then outline the significant technical challenges and open problems in system design for serving TPMs to enable scientific research and discovery. Specifically, we describe the requirements of a comprehensive software stack and interfaces to support the diverse and flexible requirements of researchers. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2402.03480v1-abstract-full').style.display = 'none'; document.getElementById('2402.03480v1-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 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">10 pages, 3 figures, accepted for publication in the proceedings of the 10th IEEE/ACM International Conference on Big Data Computing, Applications and Technologies (BDCAT2023)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2401.04552">arXiv:2401.04552</a> <span> [<a href="https://arxiv.org/pdf/2401.04552">pdf</a>, <a href="https://arxiv.org/format/2401.04552">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Distributed, Parallel, and Cluster Computing">cs.DC</span> </div> </div> <p class="title is-5 mathjax"> XaaS: Acceleration as a Service to Enable Productive High-Performance Cloud Computing </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Hoefler%2C+T">Torsten Hoefler</a>, <a href="/search/cs?searchtype=author&query=Copik%2C+M">Marcin Copik</a>, <a href="/search/cs?searchtype=author&query=Beckman%2C+P">Pete Beckman</a>, <a href="/search/cs?searchtype=author&query=Jones%2C+A">Andrew Jones</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=Parashar%2C+M">Manish Parashar</a>, <a href="/search/cs?searchtype=author&query=Reed%2C+D">Daniel Reed</a>, <a href="/search/cs?searchtype=author&query=Troyer%2C+M">Matthias Troyer</a>, <a href="/search/cs?searchtype=author&query=Schulthess%2C+T">Thomas Schulthess</a>, <a href="/search/cs?searchtype=author&query=Ernst%2C+D">Dan Ernst</a>, <a href="/search/cs?searchtype=author&query=Dongarra%2C+J">Jack Dongarra</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2401.04552v1-abstract-short" style="display: inline;"> HPC and Cloud have evolved independently, specializing their innovations into performance or productivity. Acceleration as a Service (XaaS) is a recipe to empower both fields with a shared execution platform that provides transparent access to computing resources, regardless of the underlying cloud or HPC service provider. Bridging HPC and cloud advancements, XaaS presents a unified architecture b… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2401.04552v1-abstract-full').style.display = 'inline'; document.getElementById('2401.04552v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2401.04552v1-abstract-full" style="display: none;"> HPC and Cloud have evolved independently, specializing their innovations into performance or productivity. Acceleration as a Service (XaaS) is a recipe to empower both fields with a shared execution platform that provides transparent access to computing resources, regardless of the underlying cloud or HPC service provider. Bridging HPC and cloud advancements, XaaS presents a unified architecture built on performance-portable containers. Our converged model concentrates on low-overhead, high-performance communication and computing, targeting resource-intensive workloads from climate simulations to machine learning. XaaS lifts the restricted allocation model of Function-as-a-Service (FaaS), allowing users to benefit from the flexibility and efficient resource utilization of serverless while supporting long-running and performance-sensitive workloads from HPC. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2401.04552v1-abstract-full').style.display = 'none'; document.getElementById('2401.04552v1-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 January, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> January 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2401.02524">arXiv:2401.02524</a> <span> [<a href="https://arxiv.org/pdf/2401.02524">pdf</a>, <a href="https://arxiv.org/format/2401.02524">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"> Comprehensive Exploration of Synthetic Data Generation: A Survey </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Bauer%2C+A">Andr茅 Bauer</a>, <a href="/search/cs?searchtype=author&query=Trapp%2C+S">Simon Trapp</a>, <a href="/search/cs?searchtype=author&query=Stenger%2C+M">Michael Stenger</a>, <a href="/search/cs?searchtype=author&query=Leppich%2C+R">Robert Leppich</a>, <a href="/search/cs?searchtype=author&query=Kounev%2C+S">Samuel Kounev</a>, <a href="/search/cs?searchtype=author&query=Leznik%2C+M">Mark Leznik</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2401.02524v2-abstract-short" style="display: inline;"> Recent years have witnessed a surge in the popularity of Machine Learning (ML), applied across diverse domains. However, progress is impeded by the scarcity of training data due to expensive acquisition and privacy legislation. Synthetic data emerges as a solution, but the abundance of released models and limited overview literature pose challenges for decision-making. This work surveys 417 Synthe… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2401.02524v2-abstract-full').style.display = 'inline'; document.getElementById('2401.02524v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2401.02524v2-abstract-full" style="display: none;"> Recent years have witnessed a surge in the popularity of Machine Learning (ML), applied across diverse domains. However, progress is impeded by the scarcity of training data due to expensive acquisition and privacy legislation. Synthetic data emerges as a solution, but the abundance of released models and limited overview literature pose challenges for decision-making. This work surveys 417 Synthetic Data Generation (SDG) models over the last decade, providing a comprehensive overview of model types, functionality, and improvements. Common attributes are identified, leading to a classification and trend analysis. The findings reveal increased model performance and complexity, with neural network-based approaches prevailing, except for privacy-preserving data generation. Computer vision dominates, with GANs as primary generative models, while diffusion models, transformers, and RNNs compete. Implications from our performance evaluation highlight the scarcity of common metrics and datasets, making comparisons challenging. Additionally, the neglect of training and computational costs in literature necessitates attention in future research. This work serves as a guide for SDG model selection and identifies crucial areas for future exploration. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2401.02524v2-abstract-full').style.display = 'none'; document.getElementById('2401.02524v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 1 February, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 4 January, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> January 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Fixed bug in Figure 44</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2312.10188">arXiv:2312.10188</a> <span> [<a href="https://arxiv.org/pdf/2312.10188">pdf</a>, <a href="https://arxiv.org/format/2312.10188">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"> WordScape: a Pipeline to extract multilingual, visually rich Documents with Layout Annotations from Web Crawl Data </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Weber%2C+M">Maurice Weber</a>, <a href="/search/cs?searchtype=author&query=Siebenschuh%2C+C">Carlo Siebenschuh</a>, <a href="/search/cs?searchtype=author&query=Butler%2C+R">Rory Butler</a>, <a href="/search/cs?searchtype=author&query=Alexandrov%2C+A">Anton Alexandrov</a>, <a href="/search/cs?searchtype=author&query=Thanner%2C+V">Valdemar Thanner</a>, <a href="/search/cs?searchtype=author&query=Tsolakis%2C+G">Georgios Tsolakis</a>, <a href="/search/cs?searchtype=author&query=Jabbar%2C+H">Haris Jabbar</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=Li%2C+B">Bo Li</a>, <a href="/search/cs?searchtype=author&query=Stevens%2C+R">Rick Stevens</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+C">Ce Zhang</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="2312.10188v1-abstract-short" style="display: inline;"> We introduce WordScape, a novel pipeline for the creation of cross-disciplinary, multilingual corpora comprising millions of pages with annotations for document layout detection. Relating visual and textual items on document pages has gained further significance with the advent of multimodal models. Various approaches proved effective for visual question answering or layout segmentation. However,… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.10188v1-abstract-full').style.display = 'inline'; document.getElementById('2312.10188v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2312.10188v1-abstract-full" style="display: none;"> We introduce WordScape, a novel pipeline for the creation of cross-disciplinary, multilingual corpora comprising millions of pages with annotations for document layout detection. Relating visual and textual items on document pages has gained further significance with the advent of multimodal models. Various approaches proved effective for visual question answering or layout segmentation. However, the interplay of text, tables, and visuals remains challenging for a variety of document understanding tasks. In particular, many models fail to generalize well to diverse domains and new languages due to insufficient availability of training data. WordScape addresses these limitations. Our automatic annotation pipeline parses the Open XML structure of Word documents obtained from the web, jointly providing layout-annotated document images and their textual representations. In turn, WordScape offers unique properties as it (1) leverages the ubiquity of the Word file format on the internet, (2) is readily accessible through the Common Crawl web corpus, (3) is adaptive to domain-specific documents, and (4) offers culturally and linguistically diverse document pages with natural semantic structure and high-quality text. Together with the pipeline, we will additionally release 9.5M urls to word documents which can be processed using WordScape to create a dataset of over 40M pages. Finally, we investigate the quality of text and layout annotations extracted by WordScape, assess the impact on document understanding benchmarks, and demonstrate that manual labeling costs can be substantially reduced. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.10188v1-abstract-full').style.display = 'none'; document.getElementById('2312.10188v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 15 December, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">NeurIPS 2023 Datasets and Benchmarks</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2312.03989">arXiv:2312.03989</a> <span> [<a href="https://arxiv.org/pdf/2312.03989">pdf</a>, <a href="https://arxiv.org/format/2312.03989">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="Materials Science">cond-mat.mtrl-sci</span> <span class="tag is-small is-grey 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="Data Analysis, Statistics and Probability">physics.data-an</span> </div> </div> <p class="title is-5 mathjax"> Rapid detection of rare events from in situ X-ray diffraction data using machine learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zheng%2C+W">Weijian Zheng</a>, <a href="/search/cs?searchtype=author&query=Park%2C+J">Jun-Sang Park</a>, <a href="/search/cs?searchtype=author&query=Kenesei%2C+P">Peter Kenesei</a>, <a href="/search/cs?searchtype=author&query=Ali%2C+A">Ahsan Ali</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+Z">Zhengchun Liu</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I+T">Ian T. Foster</a>, <a href="/search/cs?searchtype=author&query=Schwarz%2C+N">Nicholas Schwarz</a>, <a href="/search/cs?searchtype=author&query=Kettimuthu%2C+R">Rajkumar Kettimuthu</a>, <a href="/search/cs?searchtype=author&query=Miceli%2C+A">Antonino Miceli</a>, <a href="/search/cs?searchtype=author&query=Sharma%2C+H">Hemant Sharma</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="2312.03989v1-abstract-short" style="display: inline;"> High-energy X-ray diffraction methods can non-destructively map the 3D microstructure and associated attributes of metallic polycrystalline engineering materials in their bulk form. These methods are often combined with external stimuli such as thermo-mechanical loading to take snapshots over time of the evolving microstructure and attributes. However, the extreme data volumes and the high costs o… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.03989v1-abstract-full').style.display = 'inline'; document.getElementById('2312.03989v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2312.03989v1-abstract-full" style="display: none;"> High-energy X-ray diffraction methods can non-destructively map the 3D microstructure and associated attributes of metallic polycrystalline engineering materials in their bulk form. These methods are often combined with external stimuli such as thermo-mechanical loading to take snapshots over time of the evolving microstructure and attributes. However, the extreme data volumes and the high costs of traditional data acquisition and reduction approaches pose a barrier to quickly extracting actionable insights and improving the temporal resolution of these snapshots. Here we present a fully automated technique capable of rapidly detecting the onset of plasticity in high-energy X-ray microscopy data. Our technique is computationally faster by at least 50 times than the traditional approaches and works for data sets that are up to 9 times sparser than a full data set. This new technique leverages self-supervised image representation learning and clustering to transform massive data into compact, semantic-rich representations of visually salient characteristics (e.g., peak shapes). These characteristics can be a rapid indicator of anomalous events such as changes in diffraction peak shapes. We anticipate that this technique will provide just-in-time actionable information to drive smarter experiments that effectively deploy multi-modal X-ray diffraction methods that span many decades of length scales. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.03989v1-abstract-full').style.display = 'none'; document.getElementById('2312.03989v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 6 December, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2312.03876">arXiv:2312.03876</a> <span> [<a href="https://arxiv.org/pdf/2312.03876">pdf</a>, <a href="https://arxiv.org/format/2312.03876">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Atmospheric and Oceanic Physics">physics.ao-ph</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> </div> </div> <p class="title is-5 mathjax"> Scaling transformer neural networks for skillful and reliable medium-range weather forecasting </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Nguyen%2C+T">Tung Nguyen</a>, <a href="/search/cs?searchtype=author&query=Shah%2C+R">Rohan Shah</a>, <a href="/search/cs?searchtype=author&query=Bansal%2C+H">Hritik Bansal</a>, <a href="/search/cs?searchtype=author&query=Arcomano%2C+T">Troy Arcomano</a>, <a href="/search/cs?searchtype=author&query=Maulik%2C+R">Romit Maulik</a>, <a href="/search/cs?searchtype=author&query=Kotamarthi%2C+V">Veerabhadra Kotamarthi</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=Madireddy%2C+S">Sandeep Madireddy</a>, <a href="/search/cs?searchtype=author&query=Grover%2C+A">Aditya Grover</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="2312.03876v2-abstract-short" style="display: inline;"> Weather forecasting is a fundamental problem for anticipating and mitigating the impacts of climate change. Recently, data-driven approaches for weather forecasting based on deep learning have shown great promise, achieving accuracies that are competitive with operational systems. However, those methods often employ complex, customized architectures without sufficient ablation analysis, making it… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.03876v2-abstract-full').style.display = 'inline'; document.getElementById('2312.03876v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2312.03876v2-abstract-full" style="display: none;"> Weather forecasting is a fundamental problem for anticipating and mitigating the impacts of climate change. Recently, data-driven approaches for weather forecasting based on deep learning have shown great promise, achieving accuracies that are competitive with operational systems. However, those methods often employ complex, customized architectures without sufficient ablation analysis, making it difficult to understand what truly contributes to their success. Here we introduce Stormer, a simple transformer model that achieves state-of-the-art performance on weather forecasting with minimal changes to the standard transformer backbone. We identify the key components of Stormer through careful empirical analyses, including weather-specific embedding, randomized dynamics forecast, and pressure-weighted loss. At the core of Stormer is a randomized forecasting objective that trains the model to forecast the weather dynamics over varying time intervals. During inference, this allows us to produce multiple forecasts for a target lead time and combine them to obtain better forecast accuracy. On WeatherBench 2, Stormer performs competitively at short to medium-range forecasts and outperforms current methods beyond 7 days, while requiring orders-of-magnitude less training data and compute. Additionally, we demonstrate Stormer's favorable scaling properties, showing consistent improvements in forecast accuracy with increases in model size and training tokens. Code and checkpoints are available at https://github.com/tung-nd/stormer. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2312.03876v2-abstract-full').style.display = 'none'; document.getElementById('2312.03876v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 22 October, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 6 December, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Neural Information Processing Systems (NeurIPS 2024)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2311.00787">arXiv:2311.00787</a> <span> [<a href="https://arxiv.org/pdf/2311.00787">pdf</a>, <a href="https://arxiv.org/format/2311.00787">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Materials Science">cond-mat.mtrl-sci</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/10.1038/s41524-024-01374-8">10.1038/s41524-024-01374-8 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Accelerating Electronic Stopping Power Predictions by 10 Million Times with a Combination of Time-Dependent Density Functional Theory and Machine Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Ward%2C+L">Logan Ward</a>, <a href="/search/cs?searchtype=author&query=Blaiszik%2C+B">Ben Blaiszik</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+C">Cheng-Wei Lee</a>, <a href="/search/cs?searchtype=author&query=Martin%2C+T">Troy Martin</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=Schleife%2C+A">Andr茅 Schleife</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2311.00787v2-abstract-short" style="display: inline;"> Knowing the rate at which particle radiation releases energy in a material, the stopping power, is key to designing nuclear reactors, medical treatments, semiconductor and quantum materials, and many other technologies. While the nuclear contribution to stopping power, i.e., elastic scattering between atoms, is well understood in the literature, the route for gathering data on the electronic contr… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2311.00787v2-abstract-full').style.display = 'inline'; document.getElementById('2311.00787v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2311.00787v2-abstract-full" style="display: none;"> Knowing the rate at which particle radiation releases energy in a material, the stopping power, is key to designing nuclear reactors, medical treatments, semiconductor and quantum materials, and many other technologies. While the nuclear contribution to stopping power, i.e., elastic scattering between atoms, is well understood in the literature, the route for gathering data on the electronic contribution has for decades remained costly and reliant on many simplifying assumptions, including that materials are isotropic. We establish a method that combines time-dependent density functional theory (TDDFT) and machine learning to reduce the time to assess new materials to mere hours on a supercomputer and provides valuable data on how atomic details influence electronic stopping. Our approach uses TDDFT to compute the electronic stopping contributions to stopping power from first principles in several directions and then machine learning to interpolate to other directions at a cost of 10 million times fewer core-hours. We demonstrate the combined approach in a study of proton irradiation in aluminum and employ it to predict how the depth of maximum energy deposition, the "Bragg Peak," varies depending on incident angle -- a quantity otherwise inaccessible to modelers. The lack of any experimental information requirement makes our method applicable to most materials, and its speed makes it a prime candidate for enabling quantum-to-continuum models of radiation damage. The prospect of reusing valuable TDDFT data for training the model make our approach appealing for applications in the age of materials data science. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2311.00787v2-abstract-full').style.display = 'none'; document.getElementById('2311.00787v2-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, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 1 November, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2310.16270">arXiv:2310.16270</a> <span> [<a href="https://arxiv.org/pdf/2310.16270">pdf</a>, <a href="https://arxiv.org/format/2310.16270">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computation and Language">cs.CL</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="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"> Attention Lens: A Tool for Mechanistically Interpreting the Attention Head Information Retrieval Mechanism </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Sakarvadia%2C+M">Mansi Sakarvadia</a>, <a href="/search/cs?searchtype=author&query=Khan%2C+A">Arham Khan</a>, <a href="/search/cs?searchtype=author&query=Ajith%2C+A">Aswathy Ajith</a>, <a href="/search/cs?searchtype=author&query=Grzenda%2C+D">Daniel Grzenda</a>, <a href="/search/cs?searchtype=author&query=Hudson%2C+N">Nathaniel Hudson</a>, <a href="/search/cs?searchtype=author&query=Bauer%2C+A">Andr茅 Bauer</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</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="2310.16270v1-abstract-short" style="display: inline;"> Transformer-based Large Language Models (LLMs) are the state-of-the-art for natural language tasks. Recent work has attempted to decode, by reverse engineering the role of linear layers, the internal mechanisms by which LLMs arrive at their final predictions for text completion tasks. Yet little is known about the specific role of attention heads in producing the final token prediction. We propose… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.16270v1-abstract-full').style.display = 'inline'; document.getElementById('2310.16270v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2310.16270v1-abstract-full" style="display: none;"> Transformer-based Large Language Models (LLMs) are the state-of-the-art for natural language tasks. Recent work has attempted to decode, by reverse engineering the role of linear layers, the internal mechanisms by which LLMs arrive at their final predictions for text completion tasks. Yet little is known about the specific role of attention heads in producing the final token prediction. We propose Attention Lens, a tool that enables researchers to translate the outputs of attention heads into vocabulary tokens via learned attention-head-specific transformations called lenses. Preliminary findings from our trained lenses indicate that attention heads play highly specialized roles in language models. The code for Attention Lens is available at github.com/msakarvadia/AttentionLens. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.16270v1-abstract-full').style.display = 'none'; document.getElementById('2310.16270v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 24 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/2310.04610">arXiv:2310.04610</a> <span> [<a href="https://arxiv.org/pdf/2310.04610">pdf</a>, <a href="https://arxiv.org/format/2310.04610">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"> DeepSpeed4Science Initiative: Enabling Large-Scale Scientific Discovery through Sophisticated AI System Technologies </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Song%2C+S+L">Shuaiwen Leon Song</a>, <a href="/search/cs?searchtype=author&query=Kruft%2C+B">Bonnie Kruft</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+M">Minjia Zhang</a>, <a href="/search/cs?searchtype=author&query=Li%2C+C">Conglong Li</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+S">Shiyang Chen</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+C">Chengming Zhang</a>, <a href="/search/cs?searchtype=author&query=Tanaka%2C+M">Masahiro Tanaka</a>, <a href="/search/cs?searchtype=author&query=Wu%2C+X">Xiaoxia Wu</a>, <a href="/search/cs?searchtype=author&query=Rasley%2C+J">Jeff Rasley</a>, <a href="/search/cs?searchtype=author&query=Awan%2C+A+A">Ammar Ahmad Awan</a>, <a href="/search/cs?searchtype=author&query=Holmes%2C+C">Connor Holmes</a>, <a href="/search/cs?searchtype=author&query=Cai%2C+M">Martin Cai</a>, <a href="/search/cs?searchtype=author&query=Ghanem%2C+A">Adam Ghanem</a>, <a href="/search/cs?searchtype=author&query=Zhou%2C+Z">Zhongzhu Zhou</a>, <a href="/search/cs?searchtype=author&query=He%2C+Y">Yuxiong He</a>, <a href="/search/cs?searchtype=author&query=Luferenko%2C+P">Pete Luferenko</a>, <a href="/search/cs?searchtype=author&query=Kumar%2C+D">Divya Kumar</a>, <a href="/search/cs?searchtype=author&query=Weyn%2C+J">Jonathan Weyn</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+R">Ruixiong Zhang</a>, <a href="/search/cs?searchtype=author&query=Klocek%2C+S">Sylwester Klocek</a>, <a href="/search/cs?searchtype=author&query=Vragov%2C+V">Volodymyr Vragov</a>, <a href="/search/cs?searchtype=author&query=AlQuraishi%2C+M">Mohammed AlQuraishi</a>, <a href="/search/cs?searchtype=author&query=Ahdritz%2C+G">Gustaf Ahdritz</a>, <a href="/search/cs?searchtype=author&query=Floristean%2C+C">Christina Floristean</a>, <a href="/search/cs?searchtype=author&query=Negri%2C+C">Cristina Negri</a> , et al. (67 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="2310.04610v2-abstract-short" style="display: inline;"> In the upcoming decade, deep learning may revolutionize the natural sciences, enhancing our capacity to model and predict natural occurrences. This could herald a new era of scientific exploration, bringing significant advancements across sectors from drug development to renewable energy. To answer this call, we present DeepSpeed4Science initiative (deepspeed4science.ai) which aims to build unique… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.04610v2-abstract-full').style.display = 'inline'; document.getElementById('2310.04610v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2310.04610v2-abstract-full" style="display: none;"> In the upcoming decade, deep learning may revolutionize the natural sciences, enhancing our capacity to model and predict natural occurrences. This could herald a new era of scientific exploration, bringing significant advancements across sectors from drug development to renewable energy. To answer this call, we present DeepSpeed4Science initiative (deepspeed4science.ai) which aims to build unique capabilities through AI system technology innovations to help domain experts to unlock today's biggest science mysteries. By leveraging DeepSpeed's current technology pillars (training, inference and compression) as base technology enablers, DeepSpeed4Science will create a new set of AI system technologies tailored for accelerating scientific discoveries by addressing their unique complexity beyond the common technical approaches used for accelerating generic large language models (LLMs). In this paper, we showcase the early progress we made with DeepSpeed4Science in addressing two of the critical system challenges in structural biology research. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.04610v2-abstract-full').style.display = 'none'; document.getElementById('2310.04610v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 11 October, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 6 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/2310.00510">arXiv:2310.00510</a> <span> [<a href="https://arxiv.org/pdf/2310.00510">pdf</a>, <a href="https://arxiv.org/format/2310.00510">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Robotics">cs.RO</span> </div> </div> <p class="title is-5 mathjax"> Exploring Benchmarks for Self-Driving Labs using Color Matching </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Ginsburg%2C+T">Tobias Ginsburg</a>, <a href="/search/cs?searchtype=author&query=Hippe%2C+K">Kyle Hippe</a>, <a href="/search/cs?searchtype=author&query=Lewis%2C+R">Ryan Lewis</a>, <a href="/search/cs?searchtype=author&query=Ozgulbas%2C+D">Doga Ozgulbas</a>, <a href="/search/cs?searchtype=author&query=Cleary%2C+A">Aileen Cleary</a>, <a href="/search/cs?searchtype=author&query=Butler%2C+R">Rory Butler</a>, <a href="/search/cs?searchtype=author&query=Stone%2C+C">Casey Stone</a>, <a href="/search/cs?searchtype=author&query=Stroka%2C+A">Abraham Stroka</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</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="2310.00510v1-abstract-short" style="display: inline;"> Self Driving Labs (SDLs) that combine automation of experimental procedures with autonomous decision making are gaining popularity as a means of increasing the throughput of scientific workflows. The task of identifying quantities of supplied colored pigments that match a target color, the color matching problem, provides a simple and flexible SDL test case, as it requires experiment proposal, sam… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.00510v1-abstract-full').style.display = 'inline'; document.getElementById('2310.00510v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2310.00510v1-abstract-full" style="display: none;"> Self Driving Labs (SDLs) that combine automation of experimental procedures with autonomous decision making are gaining popularity as a means of increasing the throughput of scientific workflows. The task of identifying quantities of supplied colored pigments that match a target color, the color matching problem, provides a simple and flexible SDL test case, as it requires experiment proposal, sample creation, and sample analysis, three common components in autonomous discovery applications. We present a robotic solution to the color matching problem that allows for fully autonomous execution of a color matching protocol. Our solution leverages the WEI science factory platform to enable portability across different robotic hardware, the use of alternative optimization methods for continuous refinement, and automated publication of results for experiment tracking and post-hoc analysis. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.00510v1-abstract-full').style.display = 'none'; document.getElementById('2310.00510v1-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, 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/2309.15871">arXiv:2309.15871</a> <span> [<a href="https://arxiv.org/pdf/2309.15871">pdf</a>, <a href="https://arxiv.org/format/2309.15871">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"> Telescope: An Automated Hybrid Forecasting Approach on a Level-Playing Field </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Bauer%2C+A">Andr茅 Bauer</a>, <a href="/search/cs?searchtype=author&query=Leznik%2C+M">Mark Leznik</a>, <a href="/search/cs?searchtype=author&query=Stenger%2C+M">Michael Stenger</a>, <a href="/search/cs?searchtype=author&query=Leppich%2C+R">Robert Leppich</a>, <a href="/search/cs?searchtype=author&query=Herbst%2C+N">Nikolas Herbst</a>, <a href="/search/cs?searchtype=author&query=Kounev%2C+S">Samuel Kounev</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</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="2309.15871v1-abstract-short" style="display: inline;"> In many areas of decision-making, forecasting is an essential pillar. Consequently, many different forecasting methods have been proposed. From our experience, recently presented forecasting methods are computationally intensive, poorly automated, tailored to a particular data set, or they lack a predictable time-to-result. To this end, we introduce Telescope, a novel machine learning-based foreca… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2309.15871v1-abstract-full').style.display = 'inline'; document.getElementById('2309.15871v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2309.15871v1-abstract-full" style="display: none;"> In many areas of decision-making, forecasting is an essential pillar. Consequently, many different forecasting methods have been proposed. From our experience, recently presented forecasting methods are computationally intensive, poorly automated, tailored to a particular data set, or they lack a predictable time-to-result. To this end, we introduce Telescope, a novel machine learning-based forecasting approach that automatically retrieves relevant information from a given time series and splits it into parts, handling each of them separately. In contrast to deep learning methods, our approach doesn't require parameterization or the need to train and fit a multitude of parameters. It operates with just one time series and provides forecasts within seconds without any additional setup. Our experiments show that Telescope outperforms recent methods by providing accurate and reliable forecasts while making no assumptions about the analyzed time series. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2309.15871v1-abstract-full').style.display = 'none'; document.getElementById('2309.15871v1-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> 26 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/2309.05605">arXiv:2309.05605</a> <span> [<a href="https://arxiv.org/pdf/2309.05605">pdf</a>, <a href="https://arxiv.org/format/2309.05605">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computation and Language">cs.CL</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="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"> Memory Injections: Correcting Multi-Hop Reasoning Failures during Inference in Transformer-Based Language Models </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Sakarvadia%2C+M">Mansi Sakarvadia</a>, <a href="/search/cs?searchtype=author&query=Ajith%2C+A">Aswathy Ajith</a>, <a href="/search/cs?searchtype=author&query=Khan%2C+A">Arham Khan</a>, <a href="/search/cs?searchtype=author&query=Grzenda%2C+D">Daniel Grzenda</a>, <a href="/search/cs?searchtype=author&query=Hudson%2C+N">Nathaniel Hudson</a>, <a href="/search/cs?searchtype=author&query=Bauer%2C+A">Andr茅 Bauer</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</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="2309.05605v3-abstract-short" style="display: inline;"> Answering multi-hop reasoning questions requires retrieving and synthesizing information from diverse sources. Large Language Models (LLMs) struggle to perform such reasoning consistently. Here we propose an approach to pinpoint and rectify multi-hop reasoning failures through targeted memory injections on LLM attention heads. First, we analyze the per-layer activations of GPT-2 models in response… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2309.05605v3-abstract-full').style.display = 'inline'; document.getElementById('2309.05605v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2309.05605v3-abstract-full" style="display: none;"> Answering multi-hop reasoning questions requires retrieving and synthesizing information from diverse sources. Large Language Models (LLMs) struggle to perform such reasoning consistently. Here we propose an approach to pinpoint and rectify multi-hop reasoning failures through targeted memory injections on LLM attention heads. First, we analyze the per-layer activations of GPT-2 models in response to single and multi-hop prompts. We then propose a mechanism that allows users to inject pertinent prompt-specific information, which we refer to as "memories," at critical LLM locations during inference. By thus enabling the LLM to incorporate additional relevant information during inference, we enhance the quality of multi-hop prompt completions. We show empirically that a simple, efficient, and targeted memory injection into a key attention layer can often increase the probability of the desired next token in multi-hop tasks, by up to 424%. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2309.05605v3-abstract-full').style.display = 'none'; document.getElementById('2309.05605v3-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> 28 February, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 11 September, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 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">Oral Presentation at BlackboxNLP Workshop at EMNLP 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/2308.13701">arXiv:2308.13701</a> <span> [<a href="https://arxiv.org/pdf/2308.13701">pdf</a>, <a href="https://arxiv.org/format/2308.13701">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Distributed, Parallel, and Cluster Computing">cs.DC</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"> Linking the Dynamic PicoProbe Analytical Electron-Optical Beam Line / Microscope to Supercomputers </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Brace%2C+A">Alexander Brace</a>, <a href="/search/cs?searchtype=author&query=Vescovi%2C+R">Rafael Vescovi</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+R">Ryan Chard</a>, <a href="/search/cs?searchtype=author&query=Saint%2C+N+D">Nickolaus D. Saint</a>, <a href="/search/cs?searchtype=author&query=Ramanathan%2C+A">Arvind Ramanathan</a>, <a href="/search/cs?searchtype=author&query=Zaluzec%2C+N+J">Nestor J. Zaluzec</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2308.13701v1-abstract-short" style="display: inline;"> The Dynamic PicoProbe at Argonne National Laboratory is undergoing upgrades that will enable it to produce up to 100s of GB of data per day. While this data is highly important for both fundamental science and industrial applications, there is currently limited on-site infrastructure to handle these high-volume data streams. We address this problem by providing a software architecture capable of s… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2308.13701v1-abstract-full').style.display = 'inline'; document.getElementById('2308.13701v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2308.13701v1-abstract-full" style="display: none;"> The Dynamic PicoProbe at Argonne National Laboratory is undergoing upgrades that will enable it to produce up to 100s of GB of data per day. While this data is highly important for both fundamental science and industrial applications, there is currently limited on-site infrastructure to handle these high-volume data streams. We address this problem by providing a software architecture capable of supporting large-scale data transfers to the neighboring supercomputers at the Argonne Leadership Computing Facility. To prepare for future scientific workflows, we implement two instructive use cases for hyperspectral and spatiotemporal datasets, which include: (i) off-site data transfer, (ii) machine learning/artificial intelligence and traditional data analysis approaches, and (iii) automatic metadata extraction and cataloging of experimental results. This infrastructure supports expected workloads and also provides domain scientists the ability to reinterrogate data from past experiments to yield additional scientific value and derive new insights. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2308.13701v1-abstract-full').style.display = 'none'; document.getElementById('2308.13701v1-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, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> August 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2308.09793">arXiv:2308.09793</a> <span> [<a href="https://arxiv.org/pdf/2308.09793">pdf</a>, <a href="https://arxiv.org/format/2308.09793">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Robotics">cs.RO</span> </div> </div> <p class="title is-5 mathjax"> Towards a Modular Architecture for Science Factories </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Vescovi%2C+R">Rafael Vescovi</a>, <a href="/search/cs?searchtype=author&query=Ginsburg%2C+T">Tobias Ginsburg</a>, <a href="/search/cs?searchtype=author&query=Hippe%2C+K">Kyle Hippe</a>, <a href="/search/cs?searchtype=author&query=Ozgulbas%2C+D">Doga Ozgulbas</a>, <a href="/search/cs?searchtype=author&query=Stone%2C+C">Casey Stone</a>, <a href="/search/cs?searchtype=author&query=Stroka%2C+A">Abraham Stroka</a>, <a href="/search/cs?searchtype=author&query=Butler%2C+R">Rory Butler</a>, <a href="/search/cs?searchtype=author&query=Blaiszik%2C+B">Ben Blaiszik</a>, <a href="/search/cs?searchtype=author&query=Brettin%2C+T">Tom Brettin</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Hereld%2C+M">Mark Hereld</a>, <a href="/search/cs?searchtype=author&query=Ramanathan%2C+A">Arvind Ramanathan</a>, <a href="/search/cs?searchtype=author&query=Stevens%2C+R">Rick Stevens</a>, <a href="/search/cs?searchtype=author&query=Vriza%2C+A">Aikaterini Vriza</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+J">Jie Xu</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+Q">Qingteng Zhang</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2308.09793v2-abstract-short" style="display: inline;"> Advances in robotic automation, high-performance computing (HPC), and artificial intelligence (AI) encourage us to conceive of science factories: large, general-purpose computation- and AI-enabled self-driving laboratories (SDLs) with the generality and scale needed both to tackle large discovery problems and to support thousands of scientists. Science factories require modular hardware and softwa… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2308.09793v2-abstract-full').style.display = 'inline'; document.getElementById('2308.09793v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2308.09793v2-abstract-full" style="display: none;"> Advances in robotic automation, high-performance computing (HPC), and artificial intelligence (AI) encourage us to conceive of science factories: large, general-purpose computation- and AI-enabled self-driving laboratories (SDLs) with the generality and scale needed both to tackle large discovery problems and to support thousands of scientists. Science factories require modular hardware and software that can be replicated for scale and (re)configured to support many applications. To this end, we propose a prototype modular science factory architecture in which reconfigurable modules encapsulating scientific instruments are linked with manipulators to form workcells, that can themselves be combined to form larger assemblages, and linked with distributed computing for simulation, AI model training and inference, and related tasks. Workflows that perform sets of actions on modules can be specified, and various applications, comprising workflows plus associated computational and data manipulation steps, can be run concurrently. We report on our experiences prototyping this architecture and applying it in experiments involving 15 different robotic apparatus, five applications (one in education, two in biology, two in materials), and a variety of workflows, across four laboratories. We describe the reuse of modules, workcells, and workflows in different applications, the migration of applications between workcells, and the use of digital twins, and suggest directions for future work aimed at yet more generality and scalability. Code and data are available at https://ad-sdl.github.io/wei2023 and in the Supplementary Information <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2308.09793v2-abstract-full').style.display = 'none'; document.getElementById('2308.09793v2-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 October, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 18 August, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> August 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2307.16080">arXiv:2307.16080</a> <span> [<a href="https://arxiv.org/pdf/2307.16080">pdf</a>, <a href="https://arxiv.org/format/2307.16080">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Programming Languages">cs.PL</span> </div> </div> <p class="title is-5 mathjax"> nelli: a lightweight frontend for MLIR </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Levental%2C+M">Maksim Levental</a>, <a href="/search/cs?searchtype=author&query=Kamatar%2C+A">Alok Kamatar</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+R">Ryan Chard</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2307.16080v2-abstract-short" style="display: inline;"> Multi-Level Intermediate Representation (MLIR) is a novel compiler infrastructure that aims to provide modular and extensible components to facilitate building domain specific compilers. However, since MLIR models programs at an intermediate level of abstraction, and most extant frontends are at a very high level of abstraction, the semantics and mechanics of the fundamental transformations availa… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2307.16080v2-abstract-full').style.display = 'inline'; document.getElementById('2307.16080v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2307.16080v2-abstract-full" style="display: none;"> Multi-Level Intermediate Representation (MLIR) is a novel compiler infrastructure that aims to provide modular and extensible components to facilitate building domain specific compilers. However, since MLIR models programs at an intermediate level of abstraction, and most extant frontends are at a very high level of abstraction, the semantics and mechanics of the fundamental transformations available in MLIR are difficult to investigate and employ in and of themselves. To address these challenges, we have developed \texttt{nelli}, a lightweight, Python-embedded, domain-specific, language for generating MLIR code. \texttt{nelli} leverages existing MLIR infrastructure to develop Pythonic syntax and semantics for various MLIR features. We describe \texttt{nelli}'s design goals, discuss key details of our implementation, and demonstrate how \texttt{nelli} enables easily defining and lowering compute kernels to diverse hardware platforms. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2307.16080v2-abstract-full').style.display = 'none'; document.getElementById('2307.16080v2-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 August, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 29 July, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2307.07895">arXiv:2307.07895</a> <span> [<a href="https://arxiv.org/pdf/2307.07895">pdf</a>, <a href="https://arxiv.org/format/2307.07895">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Distributed, Parallel, and Cluster Computing">cs.DC</span> </div> <div class="is-inline-block" style="margin-left: 0.5rem"> <div class="tags has-addons"> <span class="tag is-dark is-size-7">doi</span> <span class="tag is-light is-size-7"><a class="" href="https://doi.org/10.1109/e-Science58273.2023.10254912">10.1109/e-Science58273.2023.10254912 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> PSI/J: A Portable Interface for Submitting, Monitoring, and Managing Jobs </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Hategan-Marandiuc%2C+M">Mihael Hategan-Marandiuc</a>, <a href="/search/cs?searchtype=author&query=Merzky%2C+A">Andre Merzky</a>, <a href="/search/cs?searchtype=author&query=Collier%2C+N">Nicholson Collier</a>, <a href="/search/cs?searchtype=author&query=Maheshwari%2C+K">Ketan Maheshwari</a>, <a href="/search/cs?searchtype=author&query=Ozik%2C+J">Jonathan Ozik</a>, <a href="/search/cs?searchtype=author&query=Turilli%2C+M">Matteo Turilli</a>, <a href="/search/cs?searchtype=author&query=Wilke%2C+A">Andreas Wilke</a>, <a href="/search/cs?searchtype=author&query=Wozniak%2C+J+M">Justin M. Wozniak</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=da+Silva%2C+R+F">Rafael Ferreira da Silva</a>, <a href="/search/cs?searchtype=author&query=Jha%2C+S">Shantenu Jha</a>, <a href="/search/cs?searchtype=author&query=Laney%2C+D">Daniel Laney</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2307.07895v2-abstract-short" style="display: inline;"> It is generally desirable for high-performance computing (HPC) applications to be portable between HPC systems, for example to make use of more performant hardware, make effective use of allocations, and to co-locate compute jobs with large datasets. Unfortunately, moving scientific applications between HPC systems is challenging for various reasons, most notably that HPC systems have different HP… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2307.07895v2-abstract-full').style.display = 'inline'; document.getElementById('2307.07895v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2307.07895v2-abstract-full" style="display: none;"> It is generally desirable for high-performance computing (HPC) applications to be portable between HPC systems, for example to make use of more performant hardware, make effective use of allocations, and to co-locate compute jobs with large datasets. Unfortunately, moving scientific applications between HPC systems is challenging for various reasons, most notably that HPC systems have different HPC schedulers. We introduce PSI/J, a job management abstraction API intended to simplify the construction of software components and applications that are portable over various HPC scheduler implementations. We argue that such a system is both necessary and that no viable alternative currently exists. We analyze similar notable APIs and attempt to determine the factors that influenced their evolution and adoption by the HPC community. We base the design of PSI/J on that analysis. We describe how PSI/J has been integrated in three workflow systems and one application, and also show via experiments that PSI/J imposes minimal overhead. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2307.07895v2-abstract-full').style.display = 'none'; document.getElementById('2307.07895v2-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 September, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 15 July, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2307.05416">arXiv:2307.05416</a> <span> [<a href="https://arxiv.org/pdf/2307.05416">pdf</a>, <a href="https://arxiv.org/format/2307.05416">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Distributed, Parallel, and Cluster Computing">cs.DC</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Databases">cs.DB</span> </div> </div> <p class="title is-5 mathjax"> Optimizing Scientific Data Transfer on Globus with Error-bounded Lossy Compression </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Liu%2C+Y">Yuanjian Liu</a>, <a href="/search/cs?searchtype=author&query=Di%2C+S">Sheng Di</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=Cappello%2C+F">Franck Cappello</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2307.05416v1-abstract-short" style="display: inline;"> The increasing volume and velocity of science data necessitate the frequent movement of enormous data volumes as part of routine research activities. As a result, limited wide-area bandwidth often leads to bottlenecks in research progress. However, in many cases, consuming applications (e.g., for analysis, visualization, and machine learning) can achieve acceptable performance on reduced-precision… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2307.05416v1-abstract-full').style.display = 'inline'; document.getElementById('2307.05416v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2307.05416v1-abstract-full" style="display: none;"> The increasing volume and velocity of science data necessitate the frequent movement of enormous data volumes as part of routine research activities. As a result, limited wide-area bandwidth often leads to bottlenecks in research progress. However, in many cases, consuming applications (e.g., for analysis, visualization, and machine learning) can achieve acceptable performance on reduced-precision data, and thus researchers may wish to compromise on data precision to reduce transfer and storage costs. Error-bounded lossy compression presents a promising approach as it can significantly reduce data volumes while preserving data integrity based on user-specified error bounds. In this paper, we propose a novel data transfer framework called Ocelot that integrates error-bounded lossy compression into the Globus data transfer infrastructure. We note four key contributions: (1) Ocelot is the first integration of lossy compression in Globus to significantly improve scientific data transfer performance over wide area network (WAN). (2) We propose an effective machine-learning based lossy compression quality estimation model that can predict the quality of error-bounded lossy compressors, which is fundamental to ensure that transferred data are acceptable to users. (3) We develop optimized strategies to reduce the compression time overhead, counter the compute-node waiting time, and improve transfer speed for compressed files. (4) We perform evaluations using many real-world scientific applications across different domains and distributed Globus endpoints. Our experiments show that Ocelot can improve dataset transfer performance substantially, and the quality of lossy compression (time, ratio and data distortion) can be predicted accurately for the purpose of quality assurance. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2307.05416v1-abstract-full').style.display = 'none'; document.getElementById('2307.05416v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 11 July, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2306.08695">arXiv:2306.08695</a> <span> [<a href="https://arxiv.org/pdf/2306.08695">pdf</a>, <a href="https://arxiv.org/format/2306.08695">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Materials Science">cond-mat.mtrl-sci</span> <span class="tag is-small is-grey 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/10.1038/s42004-023-01090-2">10.1038/s42004-023-01090-2 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> A generative artificial intelligence framework based on a molecular diffusion model for the design of metal-organic frameworks for carbon capture </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Park%2C+H">Hyun Park</a>, <a href="/search/cs?searchtype=author&query=Yan%2C+X">Xiaoli Yan</a>, <a href="/search/cs?searchtype=author&query=Zhu%2C+R">Ruijie Zhu</a>, <a href="/search/cs?searchtype=author&query=Huerta%2C+E+A">E. A. Huerta</a>, <a href="/search/cs?searchtype=author&query=Chaudhuri%2C+S">Santanu Chaudhuri</a>, <a href="/search/cs?searchtype=author&query=Cooper%2C+D">Donny Cooper</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=Tajkhorshid%2C+E">Emad Tajkhorshid</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="2306.08695v2-abstract-short" style="display: inline;"> Metal-organic frameworks (MOFs) exhibit great promise for CO2 capture. However, finding the best performing materials poses computational and experimental grand challenges in view of the vast chemical space of potential building blocks. Here, we introduce GHP-MOFassemble, a generative artificial intelligence (AI), high performance framework for the rational and accelerated design of MOFs with high… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2306.08695v2-abstract-full').style.display = 'inline'; document.getElementById('2306.08695v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2306.08695v2-abstract-full" style="display: none;"> Metal-organic frameworks (MOFs) exhibit great promise for CO2 capture. However, finding the best performing materials poses computational and experimental grand challenges in view of the vast chemical space of potential building blocks. Here, we introduce GHP-MOFassemble, a generative artificial intelligence (AI), high performance framework for the rational and accelerated design of MOFs with high CO2 adsorption capacity and synthesizable linkers. GHP-MOFassemble generates novel linkers, assembled with one of three pre-selected metal nodes (Cu paddlewheel, Zn paddlewheel, Zn tetramer) into MOFs in a primitive cubic topology. GHP-MOFassemble screens and validates AI-generated MOFs for uniqueness, synthesizability, structural validity, uses molecular dynamics simulations to study their stability and chemical consistency, and crystal graph neural networks and Grand Canonical Monte Carlo simulations to quantify their CO2 adsorption capacities. We present the top six AI-generated MOFs with CO2 capacities greater than 2 $m mol/g$, i.e., higher than 96.9% of structures in the hypothetical MOF dataset. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2306.08695v2-abstract-full').style.display = 'none'; document.getElementById('2306.08695v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 12 March, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 14 June, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 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">25 pages, 17 figures, 6 tables, accepted to Nature Communications Chemistry. This work was awarded the HPCwire 2023 Editors' Choice Awards for Best Use of High Performance Data Analytics \& Artificial Intelligence see https://www.hpcwire.com/2023-readers-editors-choice-data-analytics-ai/</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">ACM Class:</span> I.2 </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> Commun Chem 7, 21 (2024) </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2306.06283">arXiv:2306.06283</a> <span> [<a href="https://arxiv.org/pdf/2306.06283">pdf</a>, <a href="https://arxiv.org/format/2306.06283">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Materials Science">cond-mat.mtrl-sci</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="Chemical Physics">physics.chem-ph</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/10.1039/D3DD00113J">10.1039/D3DD00113J <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> 14 Examples of How LLMs Can Transform Materials Science and Chemistry: A Reflection on a Large Language Model Hackathon </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Jablonka%2C+K+M">Kevin Maik Jablonka</a>, <a href="/search/cs?searchtype=author&query=Ai%2C+Q">Qianxiang Ai</a>, <a href="/search/cs?searchtype=author&query=Al-Feghali%2C+A">Alexander Al-Feghali</a>, <a href="/search/cs?searchtype=author&query=Badhwar%2C+S">Shruti Badhwar</a>, <a href="/search/cs?searchtype=author&query=Bocarsly%2C+J+D">Joshua D. Bocarsly</a>, <a href="/search/cs?searchtype=author&query=Bran%2C+A+M">Andres M Bran</a>, <a href="/search/cs?searchtype=author&query=Bringuier%2C+S">Stefan Bringuier</a>, <a href="/search/cs?searchtype=author&query=Brinson%2C+L+C">L. Catherine Brinson</a>, <a href="/search/cs?searchtype=author&query=Choudhary%2C+K">Kamal Choudhary</a>, <a href="/search/cs?searchtype=author&query=Circi%2C+D">Defne Circi</a>, <a href="/search/cs?searchtype=author&query=Cox%2C+S">Sam Cox</a>, <a href="/search/cs?searchtype=author&query=de+Jong%2C+W+A">Wibe A. de Jong</a>, <a href="/search/cs?searchtype=author&query=Evans%2C+M+L">Matthew L. Evans</a>, <a href="/search/cs?searchtype=author&query=Gastellu%2C+N">Nicolas Gastellu</a>, <a href="/search/cs?searchtype=author&query=Genzling%2C+J">Jerome Genzling</a>, <a href="/search/cs?searchtype=author&query=Gil%2C+M+V">Mar铆a Victoria Gil</a>, <a href="/search/cs?searchtype=author&query=Gupta%2C+A+K">Ankur K. Gupta</a>, <a href="/search/cs?searchtype=author&query=Hong%2C+Z">Zhi Hong</a>, <a href="/search/cs?searchtype=author&query=Imran%2C+A">Alishba Imran</a>, <a href="/search/cs?searchtype=author&query=Kruschwitz%2C+S">Sabine Kruschwitz</a>, <a href="/search/cs?searchtype=author&query=Labarre%2C+A">Anne Labarre</a>, <a href="/search/cs?searchtype=author&query=L%C3%A1la%2C+J">Jakub L谩la</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+T">Tao Liu</a>, <a href="/search/cs?searchtype=author&query=Ma%2C+S">Steven Ma</a>, <a href="/search/cs?searchtype=author&query=Majumdar%2C+S">Sauradeep Majumdar</a> , et al. (28 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="2306.06283v4-abstract-short" style="display: inline;"> Large-language models (LLMs) such as GPT-4 caught the interest of many scientists. Recent studies suggested that these models could be useful in chemistry and materials science. To explore these possibilities, we organized a hackathon. This article chronicles the projects built as part of this hackathon. Participants employed LLMs for various applications, including predicting properties of mole… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2306.06283v4-abstract-full').style.display = 'inline'; document.getElementById('2306.06283v4-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2306.06283v4-abstract-full" style="display: none;"> Large-language models (LLMs) such as GPT-4 caught the interest of many scientists. Recent studies suggested that these models could be useful in chemistry and materials science. To explore these possibilities, we organized a hackathon. This article chronicles the projects built as part of this hackathon. Participants employed LLMs for various applications, including predicting properties of molecules and materials, designing novel interfaces for tools, extracting knowledge from unstructured data, and developing new educational applications. The diverse topics and the fact that working prototypes could be generated in less than two days highlight that LLMs will profoundly impact the future of our fields. The rich collection of ideas and projects also indicates that the applications of LLMs are not limited to materials science and chemistry but offer potential benefits to a wide range of scientific disciplines. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2306.06283v4-abstract-full').style.display = 'none'; document.getElementById('2306.06283v4-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 July, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 9 June, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2305.09593">arXiv:2305.09593</a> <span> [<a href="https://arxiv.org/pdf/2305.09593">pdf</a>, <a href="https://arxiv.org/format/2305.09593">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Distributed, Parallel, and Cluster Computing">cs.DC</span> </div> </div> <p class="title is-5 mathjax"> Accelerating Communications in Federated Applications with Transparent Object Proxies </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Pauloski%2C+J+G">J. Gregory Pauloski</a>, <a href="/search/cs?searchtype=author&query=Hayot-Sasson%2C+V">Valerie Hayot-Sasson</a>, <a href="/search/cs?searchtype=author&query=Ward%2C+L">Logan Ward</a>, <a href="/search/cs?searchtype=author&query=Hudson%2C+N">Nathaniel Hudson</a>, <a href="/search/cs?searchtype=author&query=Sabino%2C+C">Charlie Sabino</a>, <a href="/search/cs?searchtype=author&query=Baughman%2C+M">Matt Baughman</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2305.09593v3-abstract-short" style="display: inline;"> Advances in networks, accelerators, and cloud services encourage programmers to reconsider where to compute -- such as when fast networks make it cost-effective to compute on remote accelerators despite added latency. Workflow and cloud-hosted serverless computing frameworks can manage multi-step computations spanning federated collections of cloud, high-performance computing (HPC), and edge syste… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2305.09593v3-abstract-full').style.display = 'inline'; document.getElementById('2305.09593v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2305.09593v3-abstract-full" style="display: none;"> Advances in networks, accelerators, and cloud services encourage programmers to reconsider where to compute -- such as when fast networks make it cost-effective to compute on remote accelerators despite added latency. Workflow and cloud-hosted serverless computing frameworks can manage multi-step computations spanning federated collections of cloud, high-performance computing (HPC), and edge systems, but passing data among computational steps via cloud storage can incur high costs. Here, we overcome this obstacle with a new programming paradigm that decouples control flow from data flow by extending the pass-by-reference model to distributed applications. We describe ProxyStore, a system that implements this paradigm by providing object proxies that act as wide-area object references with just-in-time resolution. This proxy model enables data producers to communicate data unilaterally, transparently, and efficiently to both local and remote consumers. We demonstrate the benefits of this model with synthetic benchmarks and real-world scientific applications, running across various computing platforms. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2305.09593v3-abstract-full').style.display = 'none'; document.getElementById('2305.09593v3-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 August, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 16 May, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Accepted for publication at the International Conference for High Performance Computing, Networking, Storage and Analysis (SC23)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2305.03842">arXiv:2305.03842</a> <span> [<a href="https://arxiv.org/pdf/2305.03842">pdf</a>, <a href="https://arxiv.org/format/2305.03842">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Databases">cs.DB</span> </div> </div> <p class="title is-5 mathjax"> Data Station: Delegated, Trustworthy, and Auditable Computation to Enable Data-Sharing Consortia with a Data Escrow </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Xia%2C+S">Siyuan Xia</a>, <a href="/search/cs?searchtype=author&query=Zhu%2C+Z">Zhiru Zhu</a>, <a href="/search/cs?searchtype=author&query=Zhu%2C+C">Chris Zhu</a>, <a href="/search/cs?searchtype=author&query=Zhao%2C+J">Jinjin Zhao</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Elmore%2C+A+J">Aaron J. Elmore</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=Franklin%2C+M">Michael Franklin</a>, <a href="/search/cs?searchtype=author&query=Krishnan%2C+S">Sanjay Krishnan</a>, <a href="/search/cs?searchtype=author&query=Fernandez%2C+R+C">Raul Castro Fernandez</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2305.03842v1-abstract-short" style="display: inline;"> Pooling and sharing data increases and distributes its value. But since data cannot be revoked once shared, scenarios that require controlled release of data for regulatory, privacy, and legal reasons default to not sharing. Because selectively controlling what data to release is difficult, the few data-sharing consortia that exist are often built around data-sharing agreements resulting from long… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2305.03842v1-abstract-full').style.display = 'inline'; document.getElementById('2305.03842v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2305.03842v1-abstract-full" style="display: none;"> Pooling and sharing data increases and distributes its value. But since data cannot be revoked once shared, scenarios that require controlled release of data for regulatory, privacy, and legal reasons default to not sharing. Because selectively controlling what data to release is difficult, the few data-sharing consortia that exist are often built around data-sharing agreements resulting from long and tedious one-off negotiations. We introduce Data Station, a data escrow designed to enable the formation of data-sharing consortia. Data owners share data with the escrow knowing it will not be released without their consent. Data users delegate their computation to the escrow. The data escrow relies on delegated computation to execute queries without releasing the data first. Data Station leverages hardware enclaves to generate trust among participants, and exploits the centralization of data and computation to generate an audit log. We evaluate Data Station on machine learning and data-sharing applications while running on an untrusted intermediary. In addition to important qualitative advantages, we show that Data Station: i) outperforms federated learning baselines in accuracy and runtime for the machine learning application; ii) is orders of magnitude faster than alternative secure data-sharing frameworks; and iii) introduces small overhead on the critical path. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2305.03842v1-abstract-full').style.display = 'none'; document.getElementById('2305.03842v1-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 May, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2304.14982">arXiv:2304.14982</a> <span> [<a href="https://arxiv.org/pdf/2304.14982">pdf</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="Distributed, Parallel, and Cluster Computing">cs.DC</span> </div> </div> <p class="title is-5 mathjax"> Hierarchical and Decentralised Federated Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Rana%2C+O">Omer Rana</a>, <a href="/search/cs?searchtype=author&query=Spyridopoulos%2C+T">Theodoros Spyridopoulos</a>, <a href="/search/cs?searchtype=author&query=Hudson%2C+N">Nathaniel Hudson</a>, <a href="/search/cs?searchtype=author&query=Baughman%2C+M">Matt Baughman</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a>, <a href="/search/cs?searchtype=author&query=Khan%2C+A">Aftab Khan</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2304.14982v1-abstract-short" style="display: inline;"> Federated learning has shown enormous promise as a way of training ML models in distributed environments while reducing communication costs and protecting data privacy. However, the rise of complex cyber-physical systems, such as the Internet-of-Things, presents new challenges that are not met with traditional FL methods. Hierarchical Federated Learning extends the traditional FL process to enable… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2304.14982v1-abstract-full').style.display = 'inline'; document.getElementById('2304.14982v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2304.14982v1-abstract-full" style="display: none;"> Federated learning has shown enormous promise as a way of training ML models in distributed environments while reducing communication costs and protecting data privacy. However, the rise of complex cyber-physical systems, such as the Internet-of-Things, presents new challenges that are not met with traditional FL methods. Hierarchical Federated Learning extends the traditional FL process to enable more efficient model aggregation based on application needs or characteristics of the deployment environment (e.g., resource capabilities and/or network connectivity). It illustrates the benefits of balancing processing across the cloud-edge continuum. Hierarchical Federated Learning is likely to be a key enabler for a wide range of applications, such as smart farming and smart energy management, as it can improve performance and reduce costs, whilst also enabling FL workflows to be deployed in environments that are not well-suited to traditional FL. Model aggregation algorithms, software frameworks, and infrastructures will need to be designed and implemented to make such solutions accessible to researchers and engineers across a growing set of domains. H-FL also introduces a number of new challenges. For instance, there are implicit infrastructural challenges. There is also a trade-off between having generalised models and personalised models. If there exist geographical patterns for data (e.g., soil conditions in a smart farm likely are related to the geography of the region itself), then it is crucial that models used locally can consider their own locality in addition to a globally-learned model. H-FL will be crucial to future FL solutions as it can aggregate and distribute models at multiple levels to optimally serve the trade-off between locality dependence and global anomaly robustness. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2304.14982v1-abstract-full').style.display = 'none'; document.getElementById('2304.14982v1-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> 28 April, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">11 pages, 6 figures, 25 references</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">ACM Class:</span> C.2.4; I.2.11 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2303.08803">arXiv:2303.08803</a> <span> [<a href="https://arxiv.org/pdf/2303.08803">pdf</a>, <a href="https://arxiv.org/format/2303.08803">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Distributed, Parallel, and Cluster Computing">cs.DC</span> <span class="tag is-small is-grey 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/10.1109/IPDPSW59300.2023.00018">10.1109/IPDPSW59300.2023.00018 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Cloud Services Enable Efficient AI-Guided Simulation Workflows across Heterogeneous Resources </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Ward%2C+L">Logan Ward</a>, <a href="/search/cs?searchtype=author&query=Pauloski%2C+J+G">J. Gregory Pauloski</a>, <a href="/search/cs?searchtype=author&query=Hayot-Sasson%2C+V">Valerie Hayot-Sasson</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+R">Ryan Chard</a>, <a href="/search/cs?searchtype=author&query=Babuji%2C+Y">Yadu Babuji</a>, <a href="/search/cs?searchtype=author&query=Sivaraman%2C+G">Ganesh Sivaraman</a>, <a href="/search/cs?searchtype=author&query=Choudhury%2C+S">Sutanay Choudhury</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Thakur%2C+R">Rajeev Thakur</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2303.08803v1-abstract-short" style="display: inline;"> Applications that fuse machine learning and simulation can benefit from the use of multiple computing resources, with, for example, simulation codes running on highly parallel supercomputers and AI training and inference tasks on specialized accelerators. Here, we present our experiences deploying two AI-guided simulation workflows across such heterogeneous systems. A unique aspect of our approach… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2303.08803v1-abstract-full').style.display = 'inline'; document.getElementById('2303.08803v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2303.08803v1-abstract-full" style="display: none;"> Applications that fuse machine learning and simulation can benefit from the use of multiple computing resources, with, for example, simulation codes running on highly parallel supercomputers and AI training and inference tasks on specialized accelerators. Here, we present our experiences deploying two AI-guided simulation workflows across such heterogeneous systems. A unique aspect of our approach is our use of cloud-hosted management services to manage challenging aspects of cross-resource authentication and authorization, function-as-a-service (FaaS) function invocation, and data transfer. We show that these methods can achieve performance parity with systems that rely on direct connection between resources. We achieve parity by integrating the FaaS system and data transfer capabilities with a system that passes data by reference among managers and workers, and a user-configurable steering algorithm to hide data transfer latencies. We anticipate that this ease of use can enable routine use of heterogeneous resources in computational science. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2303.08803v1-abstract-full').style.display = 'none'; document.getElementById('2303.08803v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 15 March, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2302.06751">arXiv:2302.06751</a> <span> [<a href="https://arxiv.org/pdf/2302.06751">pdf</a>, <a href="https://arxiv.org/format/2302.06751">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Hardware Architecture">cs.AR</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> </div> </div> <p class="title is-5 mathjax"> OpenHLS: High-Level Synthesis for Low-Latency Deep Neural Networks for Experimental Science </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Levental%2C+M">Maksim Levental</a>, <a href="/search/cs?searchtype=author&query=Khan%2C+A">Arham Khan</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+R">Ryan Chard</a>, <a href="/search/cs?searchtype=author&query=Yoshii%2C+K">Kazutomo Yoshii</a>, <a href="/search/cs?searchtype=author&query=Chard%2C+K">Kyle Chard</a>, <a href="/search/cs?searchtype=author&query=Foster%2C+I">Ian Foster</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="2302.06751v4-abstract-short" style="display: inline;"> In many experiment-driven scientific domains, such as high-energy physics, material science, and cosmology, high data rate experiments impose hard constraints on data acquisition systems: collected data must either be indiscriminately stored for post-processing and analysis, thereby necessitating large storage capacity, or accurately filtered in real-time, thereby necessitating low-latency process… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2302.06751v4-abstract-full').style.display = 'inline'; document.getElementById('2302.06751v4-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2302.06751v4-abstract-full" style="display: none;"> In many experiment-driven scientific domains, such as high-energy physics, material science, and cosmology, high data rate experiments impose hard constraints on data acquisition systems: collected data must either be indiscriminately stored for post-processing and analysis, thereby necessitating large storage capacity, or accurately filtered in real-time, thereby necessitating low-latency processing. Deep neural networks, effective in other filtering tasks, have not been widely employed in such data acquisition systems, due to design and deployment difficulties. We present an open source, lightweight, compiler framework, without any proprietary dependencies, OpenHLS, based on high-level synthesis techniques, for translating high-level representations of deep neural networks to low-level representations, suitable for deployment to near-sensor devices such as field-programmable gate arrays. We evaluate OpenHLS on various workloads and present a case-study implementation of a deep neural network for Bragg peak detection in the context of high-energy diffraction microscopy. We show OpenHLS is able to produce an implementation of the network with a throughput 4.8 $渭$s/sample, which is approximately a 4$\times$ improvement over the existing implementation <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2302.06751v4-abstract-full').style.display = 'none'; document.getElementById('2302.06751v4-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 15 March, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 13 February, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2023. </p> </li> </ol> <nav class="pagination is-small is-centered breathe-horizontal" role="navigation" aria-label="pagination"> <a href="" class="pagination-previous is-invisible">Previous </a> <a href="/search/?searchtype=author&query=Foster%2C+I&start=50" class="pagination-next" >Next </a> <ul class="pagination-list"> <li> <a href="/search/?searchtype=author&query=Foster%2C+I&start=0" class="pagination-link is-current" aria-label="Goto page 1">1 </a> </li> <li> <a href="/search/?searchtype=author&query=Foster%2C+I&start=50" class="pagination-link " aria-label="Page 2" aria-current="page">2 </a> </li> <li> <a href="/search/?searchtype=author&query=Foster%2C+I&start=100" class="pagination-link " aria-label="Page 3" aria-current="page">3 </a> </li> </ul> </nav> <div class="is-hidden-tablet">  <span class="help" style="display: inline-block;"><a href="https://github.com/arXiv/arxiv-search/releases">Search v0.5.6 released 2020-02-24</a>  </span> </div> </div> </main> <footer> <div class="columns is-desktop" role="navigation" aria-label="Secondary">  <div class="column"> <div class="columns"> <div class="column"> <ul class="nav-spaced"> <li><a href="https://info.arxiv.org/about">About</a></li> <li><a href="https://info.arxiv.org/help">Help</a></li> </ul> </div> <div class="column"> <ul class="nav-spaced"> <li> <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512" class="icon filter-black" role="presentation"><title>contact arXiv</title><desc>Click here to contact arXiv</desc><path d="M502.3 190.8c3.9-3.1 9.7-.2 9.7 4.7V400c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V195.6c0-5 5.7-7.8 9.7-4.7 22.4 17.4 52.1 39.5 154.1 113.6 21.1 15.4 56.7 47.8 92.2 47.6 35.7.3 72-32.8 92.3-47.6 102-74.1 131.6-96.3 154-113.7zM256 320c23.2.4 56.6-29.2 73.4-41.4 132.7-96.3 142.8-104.7 173.4-128.7 5.8-4.5 9.2-11.5 9.2-18.9v-19c0-26.5-21.5-48-48-48H48C21.5 64 0 85.5 0 112v19c0 7.4 3.4 14.3 9.2 18.9 30.6 23.9 40.7 32.4 173.4 128.7 16.8 12.2 50.2 41.8 73.4 41.4z"/></svg> <a href="https://info.arxiv.org/help/contact.html"> Contact</a> </li> <li> <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512" class="icon filter-black" role="presentation"><title>subscribe to arXiv mailings</title><desc>Click here to subscribe</desc><path d="M476 3.2L12.5 270.6c-18.1 10.4-15.8 35.6 2.2 43.2L121 358.4l287.3-253.2c5.5-4.9 13.3 2.6 8.6 8.3L176 407v80.5c0 23.6 28.5 32.9 42.5 15.8L282 426l124.6 52.2c14.2 6 30.4-2.9 33-18.2l72-432C515 7.8 493.3-6.8 476 3.2z"/></svg> <a href="https://info.arxiv.org/help/subscribe"> Subscribe</a> </li> </ul> </div> </div> </div>   <div class="column"> <div class="columns"> <div class="column"> <ul class="nav-spaced"> <li><a href="https://info.arxiv.org/help/license/index.html">Copyright</a></li> <li><a href="https://info.arxiv.org/help/policies/privacy_policy.html">Privacy Policy</a></li> </ul> </div> <div class="column sorry-app-links"> <ul class="nav-spaced"> <li><a href="https://info.arxiv.org/help/web_accessibility.html">Web Accessibility Assistance</a></li> <li> <p class="help"> <a class="a11y-main-link" href="https://status.arxiv.org" target="_blank">arXiv Operational Status <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 256 512" class="icon filter-dark_grey" role="presentation"><path d="M224.3 273l-136 136c-9.4 9.4-24.6 9.4-33.9 0l-22.6-22.6c-9.4-9.4-9.4-24.6 0-33.9l96.4-96.4-96.4-96.4c-9.4-9.4-9.4-24.6 0-33.9L54.3 103c9.4-9.4 24.6-9.4 33.9 0l136 136c9.5 9.4 9.5 24.6.1 34z"/></svg></a><br> Get status notifications via <a class="is-link" href="https://subscribe.sorryapp.com/24846f03/email/new" target="_blank"><svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512" class="icon filter-black" role="presentation"><path d="M502.3 190.8c3.9-3.1 9.7-.2 9.7 4.7V400c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V195.6c0-5 5.7-7.8 9.7-4.7 22.4 17.4 52.1 39.5 154.1 113.6 21.1 15.4 56.7 47.8 92.2 47.6 35.7.3 72-32.8 92.3-47.6 102-74.1 131.6-96.3 154-113.7zM256 320c23.2.4 56.6-29.2 73.4-41.4 132.7-96.3 142.8-104.7 173.4-128.7 5.8-4.5 9.2-11.5 9.2-18.9v-19c0-26.5-21.5-48-48-48H48C21.5 64 0 85.5 0 112v19c0 7.4 3.4 14.3 9.2 18.9 30.6 23.9 40.7 32.4 173.4 128.7 16.8 12.2 50.2 41.8 73.4 41.4z"/></svg>email</a> or <a class="is-link" href="https://subscribe.sorryapp.com/24846f03/slack/new" target="_blank"><svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 448 512" class="icon filter-black" role="presentation"><path d="M94.12 315.1c0 25.9-21.16 47.06-47.06 47.06S0 341 0 315.1c0-25.9 21.16-47.06 47.06-47.06h47.06v47.06zm23.72 0c0-25.9 21.16-47.06 47.06-47.06s47.06 21.16 47.06 47.06v117.84c0 25.9-21.16 47.06-47.06 47.06s-47.06-21.16-47.06-47.06V315.1zm47.06-188.98c-25.9 0-47.06-21.16-47.06-47.06S139 32 164.9 32s47.06 21.16 47.06 47.06v47.06H164.9zm0 23.72c25.9 0 47.06 21.16 47.06 47.06s-21.16 47.06-47.06 47.06H47.06C21.16 243.96 0 222.8 0 196.9s21.16-47.06 47.06-47.06H164.9zm188.98 47.06c0-25.9 21.16-47.06 47.06-47.06 25.9 0 47.06 21.16 47.06 47.06s-21.16 47.06-47.06 47.06h-47.06V196.9zm-23.72 0c0 25.9-21.16 47.06-47.06 47.06-25.9 0-47.06-21.16-47.06-47.06V79.06c0-25.9 21.16-47.06 47.06-47.06 25.9 0 47.06 21.16 47.06 47.06V196.9zM283.1 385.88c25.9 0 47.06 21.16 47.06 47.06 0 25.9-21.16 47.06-47.06 47.06-25.9 0-47.06-21.16-47.06-47.06v-47.06h47.06zm0-23.72c-25.9 0-47.06-21.16-47.06-47.06 0-25.9 21.16-47.06 47.06-47.06h117.84c25.9 0 47.06 21.16 47.06 47.06 0 25.9-21.16 47.06-47.06 47.06H283.1z"/></svg>slack</a> </p> </li> </ul> </div> </div> </div>  </div> </footer> <script src="https://static.arxiv.org/static/base/1.0.0a5/js/member_acknowledgement.js"></script> </body> </html>

CINXE.COM

Search | arXiv e-print repository