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 52 results for author: <span class="mathjax">Thomas, G</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=Thomas%2C+G">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="Thomas, G"> </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=Thomas%2C+G&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="Thomas, G"> <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=Thomas%2C+G&start=50" class="pagination-next" >Next </a> <ul class="pagination-list"> <li> <a href="/search/?searchtype=author&query=Thomas%2C+G&start=0" class="pagination-link is-current" aria-label="Goto page 1">1 </a> </li> <li> <a href="/search/?searchtype=author&query=Thomas%2C+G&start=50" class="pagination-link " aria-label="Page 2" aria-current="page">2 </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/2502.08214">arXiv:2502.08214</a> <span> [<a href="https://arxiv.org/pdf/2502.08214">pdf</a>, <a href="https://arxiv.org/format/2502.08214">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> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Quantitative Methods">q-bio.QM</span> </div> </div> <p class="title is-5 mathjax"> Unbiased and Error-Detecting Combinatorial Pooling Experiments with Balanced Constant-Weight Gray Codes for Consecutive Positives Detection </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=He%2C+G">Guanchen He</a>, <a href="/search/cs?searchtype=author&query=Kovaleva%2C+V+A">Vasilisa A. Kovaleva</a>, <a href="/search/cs?searchtype=author&query=Barton%2C+C">Carl Barton</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+P+G">Paul G. Thomas</a>, <a href="/search/cs?searchtype=author&query=Pogorelyy%2C+M+V">Mikhail V. Pogorelyy</a>, <a href="/search/cs?searchtype=author&query=Meyer%2C+H+V">Hannah V. Meyer</a>, <a href="/search/cs?searchtype=author&query=Huang%2C+Q">Qin Huang</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="2502.08214v1-abstract-short" style="display: inline;"> Combinatorial pooling schemes have enabled the measurement of thousands of experiments in a small number of reactions. This efficiency is achieved by distributing the items to be measured across multiple reaction units called pools. However, current methods for the design of pooling schemes do not adequately address the need for balanced item distribution across pools, a property particularly impo… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2502.08214v1-abstract-full').style.display = 'inline'; document.getElementById('2502.08214v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2502.08214v1-abstract-full" style="display: none;"> Combinatorial pooling schemes have enabled the measurement of thousands of experiments in a small number of reactions. This efficiency is achieved by distributing the items to be measured across multiple reaction units called pools. However, current methods for the design of pooling schemes do not adequately address the need for balanced item distribution across pools, a property particularly important for biological applications. Here, we introduce balanced constant-weight Gray codes for detecting consecutive positives (DCP-CWGCs) for the efficient construction of combinatorial pooling schemes. Balanced DCP-CWGCs ensure uniform item distribution across pools, allow for the identification of consecutive positive items such as overlapping biological sequences, and enable error detection by keeping the number of tests on individual and consecutive positive items constant. For the efficient construction of balanced DCP-CWGCs, we have released an open-source python package codePub, with implementations of the two core algorithms: a branch-and-bound algorithm (BBA) and a recursive combination with BBA (rcBBA). Simulations using codePub show that our algorithms can construct long, balanced DCP-CWGCs that allow for error detection in tractable runtime. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2502.08214v1-abstract-full').style.display = 'none'; document.getElementById('2502.08214v1-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 February, 2025; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2025. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2502.06049">arXiv:2502.06049</a> <span> [<a href="https://arxiv.org/pdf/2502.06049">pdf</a>, <a href="https://arxiv.org/format/2502.06049">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"> LM2: Large Memory Models </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Kang%2C+J">Jikun Kang</a>, <a href="/search/cs?searchtype=author&query=Wu%2C+W">Wenqi Wu</a>, <a href="/search/cs?searchtype=author&query=Christianos%2C+F">Filippos Christianos</a>, <a href="/search/cs?searchtype=author&query=Chan%2C+A+J">Alex J. Chan</a>, <a href="/search/cs?searchtype=author&query=Greenlee%2C+F">Fraser Greenlee</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">George Thomas</a>, <a href="/search/cs?searchtype=author&query=Purtorab%2C+M">Marvin Purtorab</a>, <a href="/search/cs?searchtype=author&query=Toulis%2C+A">Andy Toulis</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="2502.06049v1-abstract-short" style="display: inline;"> This paper introduces the Large Memory Model (LM2), a decoder-only Transformer architecture enhanced with an auxiliary memory module that aims to address the limitations of standard Transformers in multi-step reasoning, relational argumentation, and synthesizing information distributed over long contexts. The proposed LM2 incorporates a memory module that acts as a contextual representation reposi… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2502.06049v1-abstract-full').style.display = 'inline'; document.getElementById('2502.06049v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2502.06049v1-abstract-full" style="display: none;"> This paper introduces the Large Memory Model (LM2), a decoder-only Transformer architecture enhanced with an auxiliary memory module that aims to address the limitations of standard Transformers in multi-step reasoning, relational argumentation, and synthesizing information distributed over long contexts. The proposed LM2 incorporates a memory module that acts as a contextual representation repository, interacting with input tokens via cross attention and updating through gating mechanisms. To preserve the Transformers general-purpose capabilities, LM2 maintains the original information flow while integrating a complementary memory pathway. Experimental results on the BABILong benchmark demonstrate that the LM2model outperforms both the memory-augmented RMT model by 37.1% and the baseline Llama-3.2 model by 86.3% on average across tasks. LM2 exhibits exceptional capabilities in multi-hop inference, numerical reasoning, and large-context question-answering. On the MMLU dataset, it achieves a 5.0% improvement over a pre-trained vanilla model, demonstrating that its memory module does not degrade performance on general tasks. Further, in our analysis, we explore the memory interpretability, effectiveness of memory modules, and test-time behavior. Our findings emphasize the importance of explicit memory in enhancing Transformer architectures. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2502.06049v1-abstract-full').style.display = 'none'; document.getElementById('2502.06049v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 9 February, 2025; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2025. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2411.09834">arXiv:2411.09834</a> <span> [<a href="https://arxiv.org/pdf/2411.09834">pdf</a>, <a href="https://arxiv.org/format/2411.09834">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"> A Benchmark for Long-Form Medical Question Answering </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Hosseini%2C+P">Pedram Hosseini</a>, <a href="/search/cs?searchtype=author&query=Sin%2C+J+M">Jessica M. Sin</a>, <a href="/search/cs?searchtype=author&query=Ren%2C+B">Bing Ren</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+B+G">Bryceton G. Thomas</a>, <a href="/search/cs?searchtype=author&query=Nouri%2C+E">Elnaz Nouri</a>, <a href="/search/cs?searchtype=author&query=Farahanchi%2C+A">Ali Farahanchi</a>, <a href="/search/cs?searchtype=author&query=Hassanpour%2C+S">Saeed Hassanpour</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="2411.09834v2-abstract-short" style="display: inline;"> There is a lack of benchmarks for evaluating large language models (LLMs) in long-form medical question answering (QA). Most existing medical QA evaluation benchmarks focus on automatic metrics and multiple-choice questions. While valuable, these benchmarks fail to fully capture or assess the complexities of real-world clinical applications where LLMs are being deployed. Furthermore, existing stud… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.09834v2-abstract-full').style.display = 'inline'; document.getElementById('2411.09834v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2411.09834v2-abstract-full" style="display: none;"> There is a lack of benchmarks for evaluating large language models (LLMs) in long-form medical question answering (QA). Most existing medical QA evaluation benchmarks focus on automatic metrics and multiple-choice questions. While valuable, these benchmarks fail to fully capture or assess the complexities of real-world clinical applications where LLMs are being deployed. Furthermore, existing studies on evaluating long-form answer generation in medical QA are primarily closed-source, lacking access to human medical expert annotations, which makes it difficult to reproduce results and enhance existing baselines. In this work, we introduce a new publicly available benchmark featuring real-world consumer medical questions with long-form answer evaluations annotated by medical doctors. We performed pairwise comparisons of responses from various open and closed-source medical and general-purpose LLMs based on criteria such as correctness, helpfulness, harmfulness, and bias. Additionally, we performed a comprehensive LLM-as-a-judge analysis to study the alignment between human judgments and LLMs. Our preliminary results highlight the strong potential of open LLMs in medical QA compared to leading closed models. Code & Data: https://github.com/lavita-ai/medical-eval-sphere <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2411.09834v2-abstract-full').style.display = 'none'; document.getElementById('2411.09834v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 19 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 14 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">AIM-FM: Advancements in Medical Foundation Models Workshop, 38th Conference on 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/2409.18999">arXiv:2409.18999</a> <span> [<a href="https://arxiv.org/pdf/2409.18999">pdf</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="Machine Learning">cs.LG</span> </div> </div> <p class="title is-5 mathjax"> Enhancing TinyBERT for Financial Sentiment Analysis Using GPT-Augmented FinBERT Distillation </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Thomas%2C+G+J">Graison Jos Thomas</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="2409.18999v1-abstract-short" style="display: inline;"> In the rapidly evolving field of financial sentiment analysis, the efficiency and accuracy of predictive models are critical due to their significant impact on financial markets. Transformer based models like BERT and large language models (LLMs) like GPT-4, have advanced NLP tasks considerably. Despite their advantages, BERT-based models face challenges with computational intensity in edge comput… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2409.18999v1-abstract-full').style.display = 'inline'; document.getElementById('2409.18999v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2409.18999v1-abstract-full" style="display: none;"> In the rapidly evolving field of financial sentiment analysis, the efficiency and accuracy of predictive models are critical due to their significant impact on financial markets. Transformer based models like BERT and large language models (LLMs) like GPT-4, have advanced NLP tasks considerably. Despite their advantages, BERT-based models face challenges with computational intensity in edge computing environments, and the substantial size and compute requirements of LLMs limit their practical deployment. This study proposes leveraging the generative capabilities of LLMs, such as GPT-4 Omni, to create synthetic, domain-specific training data. This approach addresses the challenge of data scarcity and enhances the performance of smaller models by making them competitive with their larger counterparts. The research specifically aims to enhance FinBERT, a BERT model fine-tuned for financial sentiment analysis, and develop TinyFinBERT, a compact transformer model, through a structured, two-tiered knowledge distillation strategy. Using data augmented by GPT-4 Omni, which involves generating new training examples and transforming existing data, we significantly improved the accuracy of FinBERT, preparing it to serve as a teacher model. This enhanced FinBERT then distilled knowledge to TinyFinBERT, employing both GPT-4 Omni and GPT-3.5 Turbo augmented data. The distillation strategy incorporated both logit and intermediate layer distillation. The training and evaluation of TinyFinBERT utilized the PhraseBank dataset and the FiQA 2018 Task1 dataset, achieving performance comparable to FinBERT while being substantially smaller and more efficient. This research demonstrates how LLMs can effectively contribute to the advancement of financial sentiment analysis by enhancing the capabilities of smaller, more efficient models through innovative data augmentation and distillation techniques. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2409.18999v1-abstract-full').style.display = 'none'; document.getElementById('2409.18999v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 19 September, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Submitted in partial fulfillment of the requirements for Masters in Machine Learning and Artificial Intelligence at Liverpool John Moores University, 97 pages, 1 figure, 14 tables</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">ACM Class:</span> I.2.6 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2405.07827">arXiv:2405.07827</a> <span> [<a href="https://arxiv.org/pdf/2405.07827">pdf</a>, <a href="https://arxiv.org/format/2405.07827">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Multimedia">cs.MM</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"> Automatic Recognition of Food Ingestion Environment from the AIM-2 Wearable Sensor </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Huang%2C+Y">Yuning Huang</a>, <a href="/search/cs?searchtype=author&query=Hassan%2C+M+A">Mohamed Abul Hassan</a>, <a href="/search/cs?searchtype=author&query=He%2C+J">Jiangpeng He</a>, <a href="/search/cs?searchtype=author&query=Higgins%2C+J">Janine Higgins</a>, <a href="/search/cs?searchtype=author&query=McCrory%2C+M">Megan McCrory</a>, <a href="/search/cs?searchtype=author&query=Eicher-Miller%2C+H">Heather Eicher-Miller</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Graham Thomas</a>, <a href="/search/cs?searchtype=author&query=Sazonov%2C+E+O">Edward O Sazonov</a>, <a href="/search/cs?searchtype=author&query=Zhu%2C+F+M">Fengqing Maggie Zhu</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="2405.07827v1-abstract-short" style="display: inline;"> Detecting an ingestion environment is an important aspect of monitoring dietary intake. It provides insightful information for dietary assessment. However, it is a challenging problem where human-based reviewing can be tedious, and algorithm-based review suffers from data imbalance and perceptual aliasing problems. To address these issues, we propose a neural network-based method with a two-stage… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2405.07827v1-abstract-full').style.display = 'inline'; document.getElementById('2405.07827v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2405.07827v1-abstract-full" style="display: none;"> Detecting an ingestion environment is an important aspect of monitoring dietary intake. It provides insightful information for dietary assessment. However, it is a challenging problem where human-based reviewing can be tedious, and algorithm-based review suffers from data imbalance and perceptual aliasing problems. To address these issues, we propose a neural network-based method with a two-stage training framework that tactfully combines fine-tuning and transfer learning techniques. Our method is evaluated on a newly collected dataset called ``UA Free Living Study", which uses an egocentric wearable camera, AIM-2 sensor, to simulate food consumption in free-living conditions. The proposed training framework is applied to common neural network backbones, combined with approaches in the general imbalanced classification field. Experimental results on the collected dataset show that our proposed method for automatic ingestion environment recognition successfully addresses the challenging data imbalance problem in the dataset and achieves a promising overall classification accuracy of 96.63%. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2405.07827v1-abstract-full').style.display = 'none'; document.getElementById('2405.07827v1-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 May, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Accepted at CVPRw 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/2403.19546">arXiv:2403.19546</a> <span> [<a href="https://arxiv.org/pdf/2403.19546">pdf</a>, <a href="https://arxiv.org/format/2403.19546">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="Databases">cs.DB</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Information Retrieval">cs.IR</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/10.1145/3650203.3663326">10.1145/3650203.3663326 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Croissant: A Metadata Format for ML-Ready Datasets </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Akhtar%2C+M">Mubashara Akhtar</a>, <a href="/search/cs?searchtype=author&query=Benjelloun%2C+O">Omar Benjelloun</a>, <a href="/search/cs?searchtype=author&query=Conforti%2C+C">Costanza Conforti</a>, <a href="/search/cs?searchtype=author&query=Foschini%2C+L">Luca Foschini</a>, <a href="/search/cs?searchtype=author&query=Giner-Miguelez%2C+J">Joan Giner-Miguelez</a>, <a href="/search/cs?searchtype=author&query=Gijsbers%2C+P">Pieter Gijsbers</a>, <a href="/search/cs?searchtype=author&query=Goswami%2C+S">Sujata Goswami</a>, <a href="/search/cs?searchtype=author&query=Jain%2C+N">Nitisha Jain</a>, <a href="/search/cs?searchtype=author&query=Karamousadakis%2C+M">Michalis Karamousadakis</a>, <a href="/search/cs?searchtype=author&query=Kuchnik%2C+M">Michael Kuchnik</a>, <a href="/search/cs?searchtype=author&query=Krishna%2C+S">Satyapriya Krishna</a>, <a href="/search/cs?searchtype=author&query=Lesage%2C+S">Sylvain Lesage</a>, <a href="/search/cs?searchtype=author&query=Lhoest%2C+Q">Quentin Lhoest</a>, <a href="/search/cs?searchtype=author&query=Marcenac%2C+P">Pierre Marcenac</a>, <a href="/search/cs?searchtype=author&query=Maskey%2C+M">Manil Maskey</a>, <a href="/search/cs?searchtype=author&query=Mattson%2C+P">Peter Mattson</a>, <a href="/search/cs?searchtype=author&query=Oala%2C+L">Luis Oala</a>, <a href="/search/cs?searchtype=author&query=Oderinwale%2C+H">Hamidah Oderinwale</a>, <a href="/search/cs?searchtype=author&query=Ruyssen%2C+P">Pierre Ruyssen</a>, <a href="/search/cs?searchtype=author&query=Santos%2C+T">Tim Santos</a>, <a href="/search/cs?searchtype=author&query=Shinde%2C+R">Rajat Shinde</a>, <a href="/search/cs?searchtype=author&query=Simperl%2C+E">Elena Simperl</a>, <a href="/search/cs?searchtype=author&query=Suresh%2C+A">Arjun Suresh</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Goeffry Thomas</a>, <a href="/search/cs?searchtype=author&query=Tykhonov%2C+S">Slava Tykhonov</a> , et al. (6 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="2403.19546v3-abstract-short" style="display: inline;"> Data is a critical resource for machine learning (ML), yet working with data remains a key friction point. This paper introduces Croissant, a metadata format for datasets that creates a shared representation across ML tools, frameworks, and platforms. Croissant makes datasets more discoverable, portable, and interoperable, thereby addressing significant challenges in ML data management. Croissant… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2403.19546v3-abstract-full').style.display = 'inline'; document.getElementById('2403.19546v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2403.19546v3-abstract-full" style="display: none;"> Data is a critical resource for machine learning (ML), yet working with data remains a key friction point. This paper introduces Croissant, a metadata format for datasets that creates a shared representation across ML tools, frameworks, and platforms. Croissant makes datasets more discoverable, portable, and interoperable, thereby addressing significant challenges in ML data management. Croissant is already supported by several popular dataset repositories, spanning hundreds of thousands of datasets, enabling easy loading into the most commonly-used ML frameworks, regardless of where the data is stored. Our initial evaluation by human raters shows that Croissant metadata is readable, understandable, complete, yet concise. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2403.19546v3-abstract-full').style.display = 'none'; document.getElementById('2403.19546v3-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 December, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 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">Published at the NeurIPS 2024 Datasets and Benchmark Track. A shorter version appeared earlier in Proceedings of ACM SIGMOD/PODS'24 Data Management for End-to-End Machine Learning (DEEM) Workshop https://dl.acm.org/doi/10.1145/3650203.3663326</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2311.16284">arXiv:2311.16284</a> <span> [<a href="https://arxiv.org/pdf/2311.16284">pdf</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="Human-Computer Interaction">cs.HC</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Signal Processing">eess.SP</span> </div> </div> <p class="title is-5 mathjax"> Simultaneous Energy Harvesting and Hand Gesture Recognition in Large Area Monolithic Dye-Sensitized Solar Cells </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Gethin Thomas</a>, <a href="/search/cs?searchtype=author&query=Pockett%2C+A">Adam Pockett</a>, <a href="/search/cs?searchtype=author&query=Seunarine%2C+K">Kris Seunarine</a>, <a href="/search/cs?searchtype=author&query=Carnie%2C+M">Matt Carnie</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="2311.16284v1-abstract-short" style="display: inline;"> Internet of Things (IoT) devices have become prevalent, embedding intelligence into our environment. It is projected that over 75 billion IoT devices will be connected by 2025 worldwide, with the majority being operated indoors. Dye-sensitized solar cells (DSSC) have recently been optimized for ambient light, having the capabilities of providing sufficient energy for self-powered IoT devices. Inte… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2311.16284v1-abstract-full').style.display = 'inline'; document.getElementById('2311.16284v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2311.16284v1-abstract-full" style="display: none;"> Internet of Things (IoT) devices have become prevalent, embedding intelligence into our environment. It is projected that over 75 billion IoT devices will be connected by 2025 worldwide, with the majority being operated indoors. Dye-sensitized solar cells (DSSC) have recently been optimized for ambient light, having the capabilities of providing sufficient energy for self-powered IoT devices. Interaction with digital technologies, termed Human Computer Interaction (HCI), is often achieved via physical mechanisms (e.g. remote controls, cell phones) which can hinder the natural interface between users and IoT devices, a key consideration for HCI. What if the solar cell that is powering the IoT device can also recognize hand gestures which would allow the user to naturally interact with the system? Previous attempts to achieve this have necessarily employed an array of solar cell/photodiodes to detect directionality. In this work, we demonstrate that by monitoring the photocurrent output of an asymmetrically patterned monolithic (i.e., single cell) DSSC, and using machine learning, we can recognize simple hand gestures, achieving an accuracy prediction of 97.71%. This work shows that, DSSCs are the perfect choice for self-powered interactive technologies, both in terms of powering IoT devices in ambient light conditions and having aesthetic qualities that are prioritized by users. As well as powering interactive technologies, they can also provide a means of interactive control. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2311.16284v1-abstract-full').style.display = 'none'; document.getElementById('2311.16284v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 27 November, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Main body: 10 pages, 6 figures, 3 tables. Document includes supplementary info: 30 pages, 47 supplementary 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/2310.16754">arXiv:2310.16754</a> <span> [<a href="https://arxiv.org/pdf/2310.16754">pdf</a>, <a href="https://arxiv.org/format/2310.16754">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> CAD -- Contextual Multi-modal Alignment for Dynamic AVQA </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Nadeem%2C+A">Asmar Nadeem</a>, <a href="/search/cs?searchtype=author&query=Hilton%2C+A">Adrian Hilton</a>, <a href="/search/cs?searchtype=author&query=Dawes%2C+R">Robert Dawes</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Graham Thomas</a>, <a href="/search/cs?searchtype=author&query=Mustafa%2C+A">Armin Mustafa</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="2310.16754v2-abstract-short" style="display: inline;"> In the context of Audio Visual Question Answering (AVQA) tasks, the audio visual modalities could be learnt on three levels: 1) Spatial, 2) Temporal, and 3) Semantic. Existing AVQA methods suffer from two major shortcomings; the audio-visual (AV) information passing through the network isn't aligned on Spatial and Temporal levels; and, inter-modal (audio and visual) Semantic information is often n… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.16754v2-abstract-full').style.display = 'inline'; document.getElementById('2310.16754v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2310.16754v2-abstract-full" style="display: none;"> In the context of Audio Visual Question Answering (AVQA) tasks, the audio visual modalities could be learnt on three levels: 1) Spatial, 2) Temporal, and 3) Semantic. Existing AVQA methods suffer from two major shortcomings; the audio-visual (AV) information passing through the network isn't aligned on Spatial and Temporal levels; and, inter-modal (audio and visual) Semantic information is often not balanced within a context; this results in poor performance. In this paper, we propose a novel end-to-end Contextual Multi-modal Alignment (CAD) network that addresses the challenges in AVQA methods by i) introducing a parameter-free stochastic Contextual block that ensures robust audio and visual alignment on the Spatial level; ii) proposing a pre-training technique for dynamic audio and visual alignment on Temporal level in a self-supervised setting, and iii) introducing a cross-attention mechanism to balance audio and visual information on Semantic level. The proposed novel CAD network improves the overall performance over the state-of-the-art methods on average by 9.4% on the MUSIC-AVQA dataset. We also demonstrate that our proposed contributions to AVQA can be added to the existing methods to improve their performance without additional complexity requirements. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.16754v2-abstract-full').style.display = 'none'; document.getElementById('2310.16754v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 27 October, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 25 October, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Accepted to IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) 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/2310.15582">arXiv:2310.15582</a> <span> [<a href="https://arxiv.org/pdf/2310.15582">pdf</a>, <a href="https://arxiv.org/format/2310.15582">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Cryptography and Security">cs.CR</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Programming Languages">cs.PL</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/10.1145/3590140.3629116">10.1145/3590140.3629116 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> SecV: Secure Code Partitioning via Multi-Language Secure Values </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Yuhala%2C+P">Peterson Yuhala</a>, <a href="/search/cs?searchtype=author&query=Felber%2C+P">Pascal Felber</a>, <a href="/search/cs?searchtype=author&query=Guiroux%2C+H">Hugo Guiroux</a>, <a href="/search/cs?searchtype=author&query=Lozi%2C+J">Jean-Pierre Lozi</a>, <a href="/search/cs?searchtype=author&query=Tchana%2C+A">Alain Tchana</a>, <a href="/search/cs?searchtype=author&query=Schiavoni%2C+V">Valerio Schiavoni</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Ga毛l Thomas</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="2310.15582v2-abstract-short" style="display: inline;"> Trusted execution environments like Intel SGX provide \emph{enclaves}, which offer strong security guarantees for applications. Running entire applications inside enclaves is possible, but this approach leads to a large trusted computing base (TCB). As such, various tools have been developed to partition programs written in languages such as C or Java into \emph{trusted} and \emph{untrusted} parts… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.15582v2-abstract-full').style.display = 'inline'; document.getElementById('2310.15582v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2310.15582v2-abstract-full" style="display: none;"> Trusted execution environments like Intel SGX provide \emph{enclaves}, which offer strong security guarantees for applications. Running entire applications inside enclaves is possible, but this approach leads to a large trusted computing base (TCB). As such, various tools have been developed to partition programs written in languages such as C or Java into \emph{trusted} and \emph{untrusted} parts, which are run in and out of enclaves respectively. However, those tools depend on language-specific taint-analysis and partitioning techniques. They cannot be reused for other languages and there is thus a need for tools that transcend this language barrier. We address this challenge by proposing a multi-language technique to specify sensitive code or data, as well as a multi-language tool to analyse and partition the resulting programs for trusted execution environments like Intel SGX. We leverage GraalVM's Truffle framework, which provides a language-agnostic abstract syntax tree (AST) representation for programs, to provide special AST nodes called \emph{secure nodes} that encapsulate sensitive program information. Secure nodes can easily be embedded into the ASTs of a wide range of languages via Truffle's \emph{polyglot API}. Our technique includes a multi-language dynamic taint tracking tool to analyse and partition applications based on our generic secure nodes. Our extensive evaluation with micro- and macro-benchmarks shows that we can use our technique for two languages (Javascript and \python), and that partitioned programs can obtain up to $14.5\%$ performance improvement as compared to unpartitioned versions. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.15582v2-abstract-full').style.display = 'none'; document.getElementById('2310.15582v2-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 December, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 24 October, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">12 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/2308.03901">arXiv:2308.03901</a> <span> [<a href="https://arxiv.org/pdf/2308.03901">pdf</a>, <a href="https://arxiv.org/format/2308.03901">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> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Performance">cs.PF</span> </div> </div> <p class="title is-5 mathjax"> FLIPS: Federated Learning using Intelligent Participant Selection </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Bhope%2C+R+A">Rahul Atul Bhope</a>, <a href="/search/cs?searchtype=author&query=Jayaram%2C+K+R">K. R. Jayaram</a>, <a href="/search/cs?searchtype=author&query=Venkatasubramanian%2C+N">Nalini Venkatasubramanian</a>, <a href="/search/cs?searchtype=author&query=Verma%2C+A">Ashish Verma</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Gegi Thomas</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="2308.03901v2-abstract-short" style="display: inline;"> This paper presents the design and implementation of FLIPS, a middleware system to manage data and participant heterogeneity in federated learning (FL) training workloads. In particular, we examine the benefits of label distribution clustering on participant selection in federated learning. FLIPS clusters parties involved in an FL training job based on the label distribution of their data apriori,… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2308.03901v2-abstract-full').style.display = 'inline'; document.getElementById('2308.03901v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2308.03901v2-abstract-full" style="display: none;"> This paper presents the design and implementation of FLIPS, a middleware system to manage data and participant heterogeneity in federated learning (FL) training workloads. In particular, we examine the benefits of label distribution clustering on participant selection in federated learning. FLIPS clusters parties involved in an FL training job based on the label distribution of their data apriori, and during FL training, ensures that each cluster is equitably represented in the participants selected. FLIPS can support the most common FL algorithms, including FedAvg, FedProx, FedDyn, FedOpt and FedYogi. To manage platform heterogeneity and dynamic resource availability, FLIPS incorporates a straggler management mechanism to handle changing capacities in distributed, smart community applications. Privacy of label distributions, clustering and participant selection is ensured through a trusted execution environment (TEE). Our comprehensive empirical evaluation compares FLIPS with random participant selection, as well as three other "smart" selection mechanisms - Oort, TiFL and gradient clustering using two real-world datasets, two benchmark datasets, two different non-IID distributions and three common FL algorithms (FedYogi, FedProx and FedAvg). We demonstrate that FLIPS significantly improves convergence, achieving higher accuracy by 17 - 20 % with 20 - 60 % lower communication costs, and these benefits endure in the presence of straggler participants. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2308.03901v2-abstract-full').style.display = 'none'; document.getElementById('2308.03901v2-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">v1</span> submitted 7 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/2306.10709">arXiv:2306.10709</a> <span> [<a href="https://arxiv.org/pdf/2306.10709">pdf</a>, <a href="https://arxiv.org/format/2306.10709">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computational Physics">physics.comp-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="Numerical Analysis">math.NA</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Fluid Dynamics">physics.flu-dyn</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Plasma Physics">physics.plasm-ph</span> </div> </div> <p class="title is-5 mathjax"> Machine learning of hidden variables in multiscale fluid simulation </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Joglekar%2C+A+S">Archis S. Joglekar</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+A+G+R">Alexander G. R. Thomas</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="2306.10709v1-abstract-short" style="display: inline;"> Solving fluid dynamics equations often requires the use of closure relations that account for missing microphysics. For example, when solving equations related to fluid dynamics for systems with a large Reynolds number, sub-grid effects become important and a turbulence closure is required, and in systems with a large Knudsen number, kinetic effects become important and a kinetic closure is requir… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2306.10709v1-abstract-full').style.display = 'inline'; document.getElementById('2306.10709v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2306.10709v1-abstract-full" style="display: none;"> Solving fluid dynamics equations often requires the use of closure relations that account for missing microphysics. For example, when solving equations related to fluid dynamics for systems with a large Reynolds number, sub-grid effects become important and a turbulence closure is required, and in systems with a large Knudsen number, kinetic effects become important and a kinetic closure is required. By adding an equation governing the growth and transport of the quantity requiring the closure relation, it becomes possible to capture microphysics through the introduction of ``hidden variables'' that are non-local in space and time. The behavior of the ``hidden variables'' in response to the fluid conditions can be learned from a higher fidelity or ab-initio model that contains all the microphysics. In our study, a partial differential equation simulator that is end-to-end differentiable is used to train judiciously placed neural networks against ground-truth simulations. We show that this method enables an Euler equation based approach to reproduce non-linear, large Knudsen number plasma physics that can otherwise only be modeled using Boltzmann-like equation simulators such as Vlasov or Particle-In-Cell modeling. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2306.10709v1-abstract-full').style.display = 'none'; document.getElementById('2306.10709v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 19 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/2305.02244">arXiv:2305.02244</a> <span> [<a href="https://arxiv.org/pdf/2305.02244">pdf</a>, <a href="https://arxiv.org/format/2305.02244">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Operating Systems">cs.OS</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Hardware Architecture">cs.AR</span> </div> </div> <p class="title is-5 mathjax"> NVMM cache design: Logging vs. Paging </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Dulong%2C+R">R茅mi Dulong</a>, <a href="/search/cs?searchtype=author&query=Acher%2C+Q">Quentin Acher</a>, <a href="/search/cs?searchtype=author&query=Lepers%2C+B">Baptiste Lepers</a>, <a href="/search/cs?searchtype=author&query=Schiavoni%2C+V">Valerio Schiavoni</a>, <a href="/search/cs?searchtype=author&query=Felber%2C+P">Pascal Felber</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Ga毛l Thomas</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="2305.02244v1-abstract-short" style="display: inline;"> Modern NVMM is closing the gap between DRAM and persistent storage, both in terms of performance and features. Having both byte addressability and persistence on the same device gives NVMM an unprecedented set of features, leading to the following question: How should we design an NVMM-based caching system to fully exploit its potential? We build two caching mechanisms, NVPages and NVLog, based on… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2305.02244v1-abstract-full').style.display = 'inline'; document.getElementById('2305.02244v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2305.02244v1-abstract-full" style="display: none;"> Modern NVMM is closing the gap between DRAM and persistent storage, both in terms of performance and features. Having both byte addressability and persistence on the same device gives NVMM an unprecedented set of features, leading to the following question: How should we design an NVMM-based caching system to fully exploit its potential? We build two caching mechanisms, NVPages and NVLog, based on two radically different design approaches. NVPages stores memory pages in NVMM, similar to the Linux page cache (LPC). NVLog uses NVMM to store a log of pending write operations to be submitted to the LPC, while it ensures reads with a small DRAM cache. Our study shows and quantifies advantages and flaws for both designs. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2305.02244v1-abstract-full').style.display = 'none'; document.getElementById('2305.02244v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 3 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">3 pages, 4 figures, presented for NVMW'23: 14th Annual Non-Volatile Memories Workshop</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">MSC Class:</span> 68M15 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2305.00766">arXiv:2305.00766</a> <span> [<a href="https://arxiv.org/pdf/2305.00766">pdf</a>, <a href="https://arxiv.org/format/2305.00766">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Cryptography and Security">cs.CR</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/10.1145/3464298.3493406">10.1145/3464298.3493406 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Montsalvat: Intel SGX Shielding for GraalVM Native Images </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Yuhala%2C+P">Peterson Yuhala</a>, <a href="/search/cs?searchtype=author&query=M%C3%A9n%C3%A9trey%2C+J">J盲mes M茅n茅trey</a>, <a href="/search/cs?searchtype=author&query=Felber%2C+P">Pascal Felber</a>, <a href="/search/cs?searchtype=author&query=Schiavoni%2C+V">Valerio Schiavoni</a>, <a href="/search/cs?searchtype=author&query=Tchana%2C+A">Alain Tchana</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Ga毛l Thomas</a>, <a href="/search/cs?searchtype=author&query=Guiroux%2C+H">Hugo Guiroux</a>, <a href="/search/cs?searchtype=author&query=Lozi%2C+J">Jean-Pierre Lozi</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="2305.00766v2-abstract-short" style="display: inline;"> The popularity of the Java programming language has led to its wide adoption in cloud computing infrastructures. However, Java applications running in untrusted clouds are vulnerable to various forms of privileged attacks. The emergence of trusted execution environments (TEEs) such as Intel SGX mitigates this problem. TEEs protect code and data in secure enclaves inaccessible to untrusted software… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2305.00766v2-abstract-full').style.display = 'inline'; document.getElementById('2305.00766v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2305.00766v2-abstract-full" style="display: none;"> The popularity of the Java programming language has led to its wide adoption in cloud computing infrastructures. However, Java applications running in untrusted clouds are vulnerable to various forms of privileged attacks. The emergence of trusted execution environments (TEEs) such as Intel SGX mitigates this problem. TEEs protect code and data in secure enclaves inaccessible to untrusted software, including the kernel and hypervisors. To efficiently use TEEs, developers must manually partition their applications into trusted and untrusted parts, in order to reduce the size of the trusted computing base (TCB) and minimise the risks of security vulnerabilities. However, partitioning applications poses two important challenges: (i) ensuring efficient object communication between the partitioned components, and (ii) ensuring the consistency of garbage collection between the parts, especially with memory-managed languages such as Java. We present Montsalvat, a tool which provides a practical and intuitive annotation-based partitioning approach for Java applications destined for secure enclaves. Montsalvat provides an RMI-like mechanism to ensure inter-object communication, as well as consistent garbage collection across the partitioned components. We implement Montsalvat with GraalVM native-image, a tool for compiling Java applications ahead-of-time into standalone native executables that do not require a JVM at runtime. Our extensive evaluation with micro- and macro-benchmarks shows our partitioning approach to boost performance in real-world applications <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2305.00766v2-abstract-full').style.display = 'none'; document.getElementById('2305.00766v2-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 December, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 1 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">13 pages, Proceedings of the 22nd International Middleware Conference</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2303.14829">arXiv:2303.14829</a> <span> [<a href="https://arxiv.org/pdf/2303.14829">pdf</a>, <a href="https://arxiv.org/format/2303.14829">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> SEM-POS: Grammatically and Semantically Correct Video Captioning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Nadeem%2C+A">Asmar Nadeem</a>, <a href="/search/cs?searchtype=author&query=Hilton%2C+A">Adrian Hilton</a>, <a href="/search/cs?searchtype=author&query=Dawes%2C+R">Robert Dawes</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Graham Thomas</a>, <a href="/search/cs?searchtype=author&query=Mustafa%2C+A">Armin Mustafa</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="2303.14829v2-abstract-short" style="display: inline;"> Generating grammatically and semantically correct captions in video captioning is a challenging task. The captions generated from the existing methods are either word-by-word that do not align with grammatical structure or miss key information from the input videos. To address these issues, we introduce a novel global-local fusion network, with a Global-Local Fusion Block (GLFB) that encodes and f… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2303.14829v2-abstract-full').style.display = 'inline'; document.getElementById('2303.14829v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2303.14829v2-abstract-full" style="display: none;"> Generating grammatically and semantically correct captions in video captioning is a challenging task. The captions generated from the existing methods are either word-by-word that do not align with grammatical structure or miss key information from the input videos. To address these issues, we introduce a novel global-local fusion network, with a Global-Local Fusion Block (GLFB) that encodes and fuses features from different parts of speech (POS) components with visual-spatial features. We use novel combinations of different POS components - 'determinant + subject', 'auxiliary verb', 'verb', and 'determinant + object' for supervision of the POS blocks - Det + Subject, Aux Verb, Verb, and Det + Object respectively. The novel global-local fusion network together with POS blocks helps align the visual features with language description to generate grammatically and semantically correct captions. Extensive qualitative and quantitative experiments on benchmark MSVD and MSRVTT datasets demonstrate that the proposed approach generates more grammatically and semantically correct captions compared to the existing methods, achieving the new state-of-the-art. Ablations on the POS blocks and the GLFB demonstrate the impact of the contributions on the proposed method. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2303.14829v2-abstract-full').style.display = 'none'; document.getElementById('2303.14829v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 4 April, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 26 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/2303.08789">arXiv:2303.08789</a> <span> [<a href="https://arxiv.org/pdf/2303.08789">pdf</a>, <a href="https://arxiv.org/format/2303.08789">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> <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"> PLEX: Making the Most of the Available Data for Robotic Manipulation Pretraining </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Garrett Thomas</a>, <a href="/search/cs?searchtype=author&query=Cheng%2C+C">Ching-An Cheng</a>, <a href="/search/cs?searchtype=author&query=Loynd%2C+R">Ricky Loynd</a>, <a href="/search/cs?searchtype=author&query=Frujeri%2C+F+V">Felipe Vieira Frujeri</a>, <a href="/search/cs?searchtype=author&query=Vineet%2C+V">Vibhav Vineet</a>, <a href="/search/cs?searchtype=author&query=Jalobeanu%2C+M">Mihai Jalobeanu</a>, <a href="/search/cs?searchtype=author&query=Kolobov%2C+A">Andrey Kolobov</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="2303.08789v2-abstract-short" style="display: inline;"> A rich representation is key to general robotic manipulation, but existing approaches to representation learning require large amounts of multimodal demonstrations. In this work we propose PLEX, a transformer-based architecture that learns from a small amount of task-agnostic visuomotor trajectories and a much larger amount of task-conditioned object manipulation videos -- a type of data available… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2303.08789v2-abstract-full').style.display = 'inline'; document.getElementById('2303.08789v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2303.08789v2-abstract-full" style="display: none;"> A rich representation is key to general robotic manipulation, but existing approaches to representation learning require large amounts of multimodal demonstrations. In this work we propose PLEX, a transformer-based architecture that learns from a small amount of task-agnostic visuomotor trajectories and a much larger amount of task-conditioned object manipulation videos -- a type of data available in quantity. PLEX uses visuomotor trajectories to induce a latent feature space and to learn task-agnostic manipulation routines, while diverse video-only demonstrations teach PLEX how to plan in the induced latent feature space for a wide variety of tasks. Experiments showcase PLEX's generalization on Meta-World and SOTA performance in challenging Robosuite environments. In particular, using relative positional encoding in PLEX's transformers greatly helps in low-data regimes of learning from human-collected demonstrations. The paper's accompanying code and data are available at https://microsoft.github.io/PLEX. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2303.08789v2-abstract-full').style.display = 'none'; document.getElementById('2303.08789v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 8 November, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 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/2303.00823">arXiv:2303.00823</a> <span> [<a href="https://arxiv.org/pdf/2303.00823">pdf</a>, <a href="https://arxiv.org/format/2303.00823">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Plasma Physics">physics.plasm-ph</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="Accelerator Physics">physics.acc-ph</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computational Physics">physics.comp-ph</span> </div> </div> <p class="title is-5 mathjax"> Automated control and optimisation of laser driven ion acceleration </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Loughran%2C+B">B. Loughran</a>, <a href="/search/cs?searchtype=author&query=Streeter%2C+M+J+V">M. J. V. Streeter</a>, <a href="/search/cs?searchtype=author&query=Ahmed%2C+H">H. Ahmed</a>, <a href="/search/cs?searchtype=author&query=Astbury%2C+S">S. Astbury</a>, <a href="/search/cs?searchtype=author&query=Balcazar%2C+M">M. Balcazar</a>, <a href="/search/cs?searchtype=author&query=Borghesi%2C+M">M. Borghesi</a>, <a href="/search/cs?searchtype=author&query=Bourgeois%2C+N">N. Bourgeois</a>, <a href="/search/cs?searchtype=author&query=Curry%2C+C+B">C. B. Curry</a>, <a href="/search/cs?searchtype=author&query=Dann%2C+S+J+D">S. J. D. Dann</a>, <a href="/search/cs?searchtype=author&query=DiIorio%2C+S">S. DiIorio</a>, <a href="/search/cs?searchtype=author&query=Dover%2C+N+P">N. P. Dover</a>, <a href="/search/cs?searchtype=author&query=Dzelzanis%2C+T">T. Dzelzanis</a>, <a href="/search/cs?searchtype=author&query=Ettlinger%2C+O+C">O. C. Ettlinger</a>, <a href="/search/cs?searchtype=author&query=Gauthier%2C+M">M. Gauthier</a>, <a href="/search/cs?searchtype=author&query=Giuffrida%2C+L">L. Giuffrida</a>, <a href="/search/cs?searchtype=author&query=Glenn%2C+G+D">G. D. Glenn</a>, <a href="/search/cs?searchtype=author&query=Glenzer%2C+S+H">S. H. Glenzer</a>, <a href="/search/cs?searchtype=author&query=Green%2C+J+S">J. S. Green</a>, <a href="/search/cs?searchtype=author&query=Gray%2C+R+J">R. J. Gray</a>, <a href="/search/cs?searchtype=author&query=Hicks%2C+G+S">G. S. Hicks</a>, <a href="/search/cs?searchtype=author&query=Hyland%2C+C">C. Hyland</a>, <a href="/search/cs?searchtype=author&query=Istokskaia%2C+V">V. Istokskaia</a>, <a href="/search/cs?searchtype=author&query=King%2C+M">M. King</a>, <a href="/search/cs?searchtype=author&query=Margarone%2C+D">D. Margarone</a>, <a href="/search/cs?searchtype=author&query=McCusker%2C+O">O. McCusker</a> , et al. (10 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="2303.00823v1-abstract-short" style="display: inline;"> The interaction of relativistically intense lasers with opaque targets represents a highly non-linear, multi-dimensional parameter space. This limits the utility of sequential 1D scanning of experimental parameters for the optimisation of secondary radiation, although to-date this has been the accepted methodology due to low data acquisition rates. High repetition-rate (HRR) lasers augmented by ma… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2303.00823v1-abstract-full').style.display = 'inline'; document.getElementById('2303.00823v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2303.00823v1-abstract-full" style="display: none;"> The interaction of relativistically intense lasers with opaque targets represents a highly non-linear, multi-dimensional parameter space. This limits the utility of sequential 1D scanning of experimental parameters for the optimisation of secondary radiation, although to-date this has been the accepted methodology due to low data acquisition rates. High repetition-rate (HRR) lasers augmented by machine learning present a valuable opportunity for efficient source optimisation. Here, an automated, HRR-compatible system produced high fidelity parameter scans, revealing the influence of laser intensity on target pre-heating and proton generation. A closed-loop Bayesian optimisation of maximum proton energy, through control of the laser wavefront and target position, produced proton beams with equivalent maximum energy to manually-optimized laser pulses but using only 60% of the laser energy. This demonstration of automated optimisation of laser-driven proton beams is a crucial step towards deeper physical insight and the construction of future radiation sources. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2303.00823v1-abstract-full').style.display = 'none'; document.getElementById('2303.00823v1-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 March, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">11 pages</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2208.09740">arXiv:2208.09740</a> <span> [<a href="https://arxiv.org/pdf/2208.09740">pdf</a>, <a href="https://arxiv.org/format/2208.09740">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"> Just-in-Time Aggregation for Federated Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Jayaram%2C+K+R">K. R. Jayaram</a>, <a href="/search/cs?searchtype=author&query=Verma%2C+A">Ashish Verma</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Gegi Thomas</a>, <a href="/search/cs?searchtype=author&query=Muthusamy%2C+V">Vinod Muthusamy</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="2208.09740v1-abstract-short" style="display: inline;"> The increasing number and scale of federated learning (FL) jobs necessitates resource efficient scheduling and management of aggregation to make the economics of cloud-hosted aggregation work. Existing FL research has focused on the design of FL algorithms and optimization, and less on the efficacy of aggregation. Existing FL platforms often employ aggregators that actively wait for model updates.… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2208.09740v1-abstract-full').style.display = 'inline'; document.getElementById('2208.09740v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2208.09740v1-abstract-full" style="display: none;"> The increasing number and scale of federated learning (FL) jobs necessitates resource efficient scheduling and management of aggregation to make the economics of cloud-hosted aggregation work. Existing FL research has focused on the design of FL algorithms and optimization, and less on the efficacy of aggregation. Existing FL platforms often employ aggregators that actively wait for model updates. This wastes computational resources on the cloud, especially in large scale FL settings where parties are intermittently available for training. In this paper, we propose a new FL aggregation paradigm -- "just-in-time" (JIT) aggregation that leverages unique properties of FL jobs, especially the periodicity of model updates, to defer aggregation as much as possible and free compute resources for other FL jobs or other datacenter workloads. We describe a novel way to prioritize FL jobs for aggregation, and demonstrate using multiple datasets, models and FL aggregation algorithms that our techniques can reduce resource usage by 60+\% when compared to eager aggregation used in existing FL platforms. We also demonstrate that using JIT aggregation has negligible overhead and impact on the latency of the FL job. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2208.09740v1-abstract-full').style.display = 'none'; document.getElementById('2208.09740v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 20 August, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> August 2022. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">10 pages. Extended version of the paper accepted to MASCOTS 2022. arXiv admin note: text overlap with arXiv:2203.12163</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2206.01637">arXiv:2206.01637</a> <span> [<a href="https://arxiv.org/pdf/2206.01637">pdf</a>, <a href="https://arxiv.org/format/2206.01637">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Plasma Physics">physics.plasm-ph</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="Computational Physics">physics.comp-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/10.1017/S0022377822000939">10.1017/S0022377822000939 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Unsupervised Discovery of Inertial-Fusion Plasma Physics using Differentiable Kinetic Simulations and a Maximum Entropy Loss Function </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Joglekar%2C+A+S">Archis S. Joglekar</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+A+G+R">Alexander G. R. Thomas</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="2206.01637v2-abstract-short" style="display: inline;"> Plasma supports collective modes and particle-wave interactions that leads to complex behavior in inertial fusion energy applications. While plasma can sometimes be modeled as a charged fluid, a kinetic description is useful towards the study of nonlinear effects in the higher dimensional momentum-position phase-space that describes the full complexity of plasma dynamics. We create a differentiabl… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2206.01637v2-abstract-full').style.display = 'inline'; document.getElementById('2206.01637v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2206.01637v2-abstract-full" style="display: none;"> Plasma supports collective modes and particle-wave interactions that leads to complex behavior in inertial fusion energy applications. While plasma can sometimes be modeled as a charged fluid, a kinetic description is useful towards the study of nonlinear effects in the higher dimensional momentum-position phase-space that describes the full complexity of plasma dynamics. We create a differentiable solver for the plasma kinetics 3D partial-differential-equation and introduce a domain-specific objective function. Using this framework, we perform gradient-based optimization of neural networks that provide forcing function parameters to the differentiable solver given a set of initial conditions. We apply this to an inertial-fusion relevant configuration and find that the optimization process exploits a novel physical effect that has previously remained undiscovered. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2206.01637v2-abstract-full').style.display = 'none'; document.getElementById('2206.01637v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 27 July, 2022; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 3 June, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2022. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">2nd AI4Science Workshop at the 39th International Conference on Machine Learning (ICML), 2022</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2205.00155">arXiv:2205.00155</a> <span> [<a href="https://arxiv.org/pdf/2205.00155">pdf</a>, <a href="https://arxiv.org/ps/2205.00155">ps</a>, <a href="https://arxiv.org/format/2205.00155">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 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/10.48550/arXiv.2205.00155">10.48550/arXiv.2205.00155 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Real-Time Gait Phase and Task Estimation for Controlling a Powered Ankle Exoskeleton on Extremely Uneven Terrain </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Medrano%2C+R+L">Roberto Leo Medrano</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G+C">Gray Cortright Thomas</a>, <a href="/search/cs?searchtype=author&query=Keais%2C+C+G">Connor G. Keais</a>, <a href="/search/cs?searchtype=author&query=Rouse%2C+E+J">Elliott J. Rouse</a>, <a href="/search/cs?searchtype=author&query=Gregg%2C+R+D">Robert D. Gregg</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2205.00155v3-abstract-short" style="display: inline;"> Positive biomechanical outcomes have been reported with lower-limb exoskeletons in laboratory settings, but these devices have difficulty delivering appropriate assistance in synchrony with human gait as the task or rate of phase progression change in real-world environments. This paper presents a controller for an ankle exoskeleton that uses a data-driven kinematic model to continuously estimate… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2205.00155v3-abstract-full').style.display = 'inline'; document.getElementById('2205.00155v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2205.00155v3-abstract-full" style="display: none;"> Positive biomechanical outcomes have been reported with lower-limb exoskeletons in laboratory settings, but these devices have difficulty delivering appropriate assistance in synchrony with human gait as the task or rate of phase progression change in real-world environments. This paper presents a controller for an ankle exoskeleton that uses a data-driven kinematic model to continuously estimate the phase, phase rate, stride length, and ground incline states during locomotion, which enables the real-time adaptation of torque assistance to match human torques observed in a multi-activity database of 10 able-bodied subjects. We demonstrate in live experiments with a new cohort of 10 able-bodied participants that the controller yields phase estimates comparable to the state of the art, while also estimating task variables with similar accuracy to recent machine learning approaches. The implemented controller successfully adapts its assistance in response to changing phase and task variables, both during controlled treadmill trials (N=10, phase RMSE: 4.8 +- 2.4\%) and a real-world stress test with extremely uneven terrain (N=1, phase RMSE: 4.8 +- 2.7\%). <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2205.00155v3-abstract-full').style.display = 'none'; document.getElementById('2205.00155v3-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 October, 2022; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 30 April, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 2022. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2203.12163">arXiv:2203.12163</a> <span> [<a href="https://arxiv.org/pdf/2203.12163">pdf</a>, <a href="https://arxiv.org/format/2203.12163">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"> Adaptive Aggregation For Federated Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Jayaram%2C+K+R">K. R. Jayaram</a>, <a href="/search/cs?searchtype=author&query=Muthusamy%2C+V">Vinod Muthusamy</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Gegi Thomas</a>, <a href="/search/cs?searchtype=author&query=Verma%2C+A">Ashish Verma</a>, <a href="/search/cs?searchtype=author&query=Purcell%2C+M">Mark Purcell</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="2203.12163v2-abstract-short" style="display: inline;"> Advances in federated learning (FL) algorithms,along with technologies like differential privacy and homomorphic encryption, have led to FL being increasingly adopted and used in many application domains. This increasing adoption has led to rapid growth in the number, size (number of participants/parties) and diversity (intermittent vs. active parties) of FL jobs. Many existing FL systems, based o… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2203.12163v2-abstract-full').style.display = 'inline'; document.getElementById('2203.12163v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2203.12163v2-abstract-full" style="display: none;"> Advances in federated learning (FL) algorithms,along with technologies like differential privacy and homomorphic encryption, have led to FL being increasingly adopted and used in many application domains. This increasing adoption has led to rapid growth in the number, size (number of participants/parties) and diversity (intermittent vs. active parties) of FL jobs. Many existing FL systems, based on centralized (often single) model aggregators are unable to scale to handle large FL jobs and adapt to parties' behavior. In this paper, we present a new scalable and adaptive architecture for FL aggregation. First, we demonstrate how traditional tree overlay based aggregation techniques (from P2P, publish-subscribe and stream processing research) can help FL aggregation scale, but are ineffective from a resource utilization and cost standpoint. Next, we present the design and implementation of AdaFed, which uses serverless/cloud functions to adaptively scale aggregation in a resource efficient and fault tolerant manner. We describe how AdaFed enables FL aggregation to be dynamically deployed only when necessary, elastically scaled to handle participant joins/leaves and is fault tolerant with minimal effort required on the (aggregation) programmer side. We also demonstrate that our prototype based on Ray scales to thousands of participants, and is able to achieve a >90% reduction in resource requirements and cost, with minimal impact on aggregation latency. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2203.12163v2-abstract-full').style.display = 'none'; document.getElementById('2203.12163v2-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, 2022; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 22 March, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 2022. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">ACM Class:</span> C.2.4; C.4 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2202.12246">arXiv:2202.12246</a> <span> [<a href="https://arxiv.org/pdf/2202.12246">pdf</a>, <a href="https://arxiv.org/format/2202.12246">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> </div> </div> <p class="title is-5 mathjax"> Neural reality of argument structure constructions </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Li%2C+B">Bai Li</a>, <a href="/search/cs?searchtype=author&query=Zhu%2C+Z">Zining Zhu</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Guillaume Thomas</a>, <a href="/search/cs?searchtype=author&query=Rudzicz%2C+F">Frank Rudzicz</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+Y">Yang Xu</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2202.12246v1-abstract-short" style="display: inline;"> In lexicalist linguistic theories, argument structure is assumed to be predictable from the meaning of verbs. As a result, the verb is the primary determinant of the meaning of a clause. In contrast, construction grammarians propose that argument structure is encoded in constructions (or form-meaning pairs) that are distinct from verbs. Decades of psycholinguistic research have produced substantia… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2202.12246v1-abstract-full').style.display = 'inline'; document.getElementById('2202.12246v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2202.12246v1-abstract-full" style="display: none;"> In lexicalist linguistic theories, argument structure is assumed to be predictable from the meaning of verbs. As a result, the verb is the primary determinant of the meaning of a clause. In contrast, construction grammarians propose that argument structure is encoded in constructions (or form-meaning pairs) that are distinct from verbs. Decades of psycholinguistic research have produced substantial empirical evidence in favor of the construction view. Here we adapt several psycholinguistic studies to probe for the existence of argument structure constructions (ASCs) in Transformer-based language models (LMs). First, using a sentence sorting experiment, we find that sentences sharing the same construction are closer in embedding space than sentences sharing the same verb. Furthermore, LMs increasingly prefer grouping by construction with more input data, mirroring the behaviour of non-native language learners. Second, in a "Jabberwocky" priming-based experiment, we find that LMs associate ASCs with meaning, even in semantically nonsensical sentences. Our work offers the first evidence for ASCs in LMs and highlights the potential to devise novel probing methods grounded in psycholinguistic research. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2202.12246v1-abstract-full').style.display = 'none'; document.getElementById('2202.12246v1-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 February, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2022. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">ACL 2022 (Long Paper)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2202.07789">arXiv:2202.07789</a> <span> [<a href="https://arxiv.org/pdf/2202.07789">pdf</a>, <a href="https://arxiv.org/format/2202.07789">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"> Safe Reinforcement Learning by Imagining the Near Future </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Garrett Thomas</a>, <a href="/search/cs?searchtype=author&query=Luo%2C+Y">Yuping Luo</a>, <a href="/search/cs?searchtype=author&query=Ma%2C+T">Tengyu Ma</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2202.07789v1-abstract-short" style="display: inline;"> Safe reinforcement learning is a promising path toward applying reinforcement learning algorithms to real-world problems, where suboptimal behaviors may lead to actual negative consequences. In this work, we focus on the setting where unsafe states can be avoided by planning ahead a short time into the future. In this setting, a model-based agent with a sufficiently accurate model can avoid unsafe… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2202.07789v1-abstract-full').style.display = 'inline'; document.getElementById('2202.07789v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2202.07789v1-abstract-full" style="display: none;"> Safe reinforcement learning is a promising path toward applying reinforcement learning algorithms to real-world problems, where suboptimal behaviors may lead to actual negative consequences. In this work, we focus on the setting where unsafe states can be avoided by planning ahead a short time into the future. In this setting, a model-based agent with a sufficiently accurate model can avoid unsafe states. We devise a model-based algorithm that heavily penalizes unsafe trajectories, and derive guarantees that our algorithm can avoid unsafe states under certain assumptions. Experiments demonstrate that our algorithm can achieve competitive rewards with fewer safety violations in several continuous control tasks. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2202.07789v1-abstract-full').style.display = 'none'; document.getElementById('2202.07789v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 15 February, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2022. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Accepted at NeurIPS 2021</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2110.01562">arXiv:2110.01562</a> <span> [<a href="https://arxiv.org/pdf/2110.01562">pdf</a>, <a href="https://arxiv.org/format/2110.01562">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"> Enhancing Voluntary Motion with Modular, Backdrivable, Powered Hip and Knee Orthoses </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Nesler%2C+C">Christopher Nesler</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Gray Thomas</a>, <a href="/search/cs?searchtype=author&query=Divekar%2C+N">Nikhil Divekar</a>, <a href="/search/cs?searchtype=author&query=Rouse%2C+E+J">Elliott J. Rouse</a>, <a href="/search/cs?searchtype=author&query=Gregg%2C+R+D">Robert D. Gregg</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="2110.01562v1-abstract-short" style="display: inline;"> Mobility disabilities are prominent in society with wide-ranging detriments to affected individuals. Addressing the specific deficits of individuals within this heterogeneous population requires modular, partial-assist, lower-limb exoskeletons. This paper introduces the Modular Backdrivable Lower-limb Unloading Exoskeleton (M-BLUE), which implements high torque, low mechanical impedance actuators… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2110.01562v1-abstract-full').style.display = 'inline'; document.getElementById('2110.01562v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2110.01562v1-abstract-full" style="display: none;"> Mobility disabilities are prominent in society with wide-ranging detriments to affected individuals. Addressing the specific deficits of individuals within this heterogeneous population requires modular, partial-assist, lower-limb exoskeletons. This paper introduces the Modular Backdrivable Lower-limb Unloading Exoskeleton (M-BLUE), which implements high torque, low mechanical impedance actuators on commercial orthoses with sheet metal modifications to produce a variety of hip- and/or knee-assisting configurations. Benchtop system identification verifies the desirable backdrive properties of the actuator, and allows for torque prediction within 0.4 Nm. An able-bodied human subject experiment demonstrates that three unilateral configurations of M-BLUE (hip only, knee only, and hip-knee) with a simple gravity compensation controller can reduce muscle EMG readings in a lifting and lowering task relative to the bare condition. Reductions in mean muscular effort and peak muscle activation were seen across the primary squat musculature (excluding biceps femoris), demonstrating the potential to reduce fatigue leading to poor lifting posture. These promising results motivate applications of M-BLUE to additional subject populations such as hip/knee osteoarthritis and geriatric frailty, and the expansion of M-BLUE to bilateral and ankle configurations. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2110.01562v1-abstract-full').style.display = 'none'; document.getElementById('2110.01562v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 4 October, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2021. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">8 pages, 7 figures</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2109.02145">arXiv:2109.02145</a> <span> [<a href="https://arxiv.org/pdf/2109.02145">pdf</a>, <a href="https://arxiv.org/format/2109.02145">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"> Temporal Shift Reinforcement Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Thomas%2C+D+G">Deepak George Thomas</a>, <a href="/search/cs?searchtype=author&query=Wongpiromsarn%2C+T">Tichakorn Wongpiromsarn</a>, <a href="/search/cs?searchtype=author&query=Jannesari%2C+A">Ali Jannesari</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="2109.02145v3-abstract-short" style="display: inline;"> The function approximators employed by traditional image-based Deep Reinforcement Learning (DRL) algorithms usually lack a temporal learning component and instead focus on learning the spatial component. We propose a technique, Temporal Shift Reinforcement Learning (TSRL), wherein both temporal, as well as spatial components are jointly learned. Moreover, TSRL does not require additional parameter… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2109.02145v3-abstract-full').style.display = 'inline'; document.getElementById('2109.02145v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2109.02145v3-abstract-full" style="display: none;"> The function approximators employed by traditional image-based Deep Reinforcement Learning (DRL) algorithms usually lack a temporal learning component and instead focus on learning the spatial component. We propose a technique, Temporal Shift Reinforcement Learning (TSRL), wherein both temporal, as well as spatial components are jointly learned. Moreover, TSRL does not require additional parameters to perform temporal learning. We show that TSRL outperforms the commonly used frame stacking heuristic on both of the Atari environments we test on while beating the SOTA for one of them. This investigation has implications in the robotics as well as sequential decision-making domains. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2109.02145v3-abstract-full').style.display = 'none'; document.getElementById('2109.02145v3-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 October, 2021; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 5 September, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 2021. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2105.10397">arXiv:2105.10397</a> <span> [<a href="https://arxiv.org/pdf/2105.10397">pdf</a>, <a href="https://arxiv.org/format/2105.10397">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="Hardware Architecture">cs.AR</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Operating Systems">cs.OS</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/DSN48987.2021.00033">10.1109/DSN48987.2021.00033 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> NVCache: A Plug-and-Play NVMM-based I/O Booster for Legacy Systems </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Dulong%2C+R">R茅mi Dulong</a>, <a href="/search/cs?searchtype=author&query=Pires%2C+R">Rafael Pires</a>, <a href="/search/cs?searchtype=author&query=Correia%2C+A">Andreia Correia</a>, <a href="/search/cs?searchtype=author&query=Schiavoni%2C+V">Valerio Schiavoni</a>, <a href="/search/cs?searchtype=author&query=Ramalhete%2C+P">Pedro Ramalhete</a>, <a href="/search/cs?searchtype=author&query=Felber%2C+P">Pascal Felber</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Ga毛l Thomas</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="2105.10397v2-abstract-short" style="display: inline;"> This paper introduces NVCache, an approach that uses a non-volatile main memory (NVMM) as a write cache to improve the write performance of legacy applications. We compare NVCache against file systems tailored for NVMM (Ext4-DAX and NOVA) and with I/O-heavy applications (SQLite, RocksDB). Our evaluation shows that NVCache reaches the performance level of the existing state-of-the-art systems for N… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2105.10397v2-abstract-full').style.display = 'inline'; document.getElementById('2105.10397v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2105.10397v2-abstract-full" style="display: none;"> This paper introduces NVCache, an approach that uses a non-volatile main memory (NVMM) as a write cache to improve the write performance of legacy applications. We compare NVCache against file systems tailored for NVMM (Ext4-DAX and NOVA) and with I/O-heavy applications (SQLite, RocksDB). Our evaluation shows that NVCache reaches the performance level of the existing state-of-the-art systems for NVMM, but without their limitations: NVCache does not limit the size of the stored data to the size of the NVMM, and works transparently with unmodified legacy applications, providing additional persistence guarantees even when their source code is not available. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2105.10397v2-abstract-full').style.display = 'none'; document.getElementById('2105.10397v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 3 September, 2021; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 14 May, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 2021. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">13 pages, 7 figures, to be published in the 51th IEEE/IFIP International Conference on Dependable Systems and Networks (DSN 21)</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">MSC Class:</span> 68M20 <span class="has-text-black-bis has-text-weight-semibold">ACM Class:</span> D.4.2; D.4.3; D.4.8 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2105.07452">arXiv:2105.07452</a> <span> [<a href="https://arxiv.org/pdf/2105.07452">pdf</a>, <a href="https://arxiv.org/format/2105.07452">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> </div> </div> <p class="title is-5 mathjax"> How is BERT surprised? Layerwise detection of linguistic anomalies </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Li%2C+B">Bai Li</a>, <a href="/search/cs?searchtype=author&query=Zhu%2C+Z">Zining Zhu</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Guillaume Thomas</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+Y">Yang Xu</a>, <a href="/search/cs?searchtype=author&query=Rudzicz%2C+F">Frank Rudzicz</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="2105.07452v1-abstract-short" style="display: inline;"> Transformer language models have shown remarkable ability in detecting when a word is anomalous in context, but likelihood scores offer no information about the cause of the anomaly. In this work, we use Gaussian models for density estimation at intermediate layers of three language models (BERT, RoBERTa, and XLNet), and evaluate our method on BLiMP, a grammaticality judgement benchmark. In lower… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2105.07452v1-abstract-full').style.display = 'inline'; document.getElementById('2105.07452v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2105.07452v1-abstract-full" style="display: none;"> Transformer language models have shown remarkable ability in detecting when a word is anomalous in context, but likelihood scores offer no information about the cause of the anomaly. In this work, we use Gaussian models for density estimation at intermediate layers of three language models (BERT, RoBERTa, and XLNet), and evaluate our method on BLiMP, a grammaticality judgement benchmark. In lower layers, surprisal is highly correlated to low token frequency, but this correlation diminishes in upper layers. Next, we gather datasets of morphosyntactic, semantic, and commonsense anomalies from psycholinguistic studies; we find that the best performing model RoBERTa exhibits surprisal in earlier layers when the anomaly is morphosyntactic than when it is semantic, while commonsense anomalies do not exhibit surprisal at any intermediate layer. These results suggest that language models employ separate mechanisms to detect different types of linguistic anomalies. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2105.07452v1-abstract-full').style.display = 'none'; document.getElementById('2105.07452v1-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 May, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 2021. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">ACL 2021 (Long Paper)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2104.04060">arXiv:2104.04060</a> <span> [<a href="https://arxiv.org/pdf/2104.04060">pdf</a>, <a href="https://arxiv.org/format/2104.04060">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="Networking and Internet Architecture">cs.NI</span> </div> </div> <p class="title is-5 mathjax"> Network in Disaggregated Datacenters </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Ekane%2C+B">Brice Ekane</a>, <a href="/search/cs?searchtype=author&query=Pipereau%2C+Y">Yohan Pipereau</a>, <a href="/search/cs?searchtype=author&query=Teabe%2C+B">Boris Teabe</a>, <a href="/search/cs?searchtype=author&query=Tchana%2C+A">Alain Tchana</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Gael Thomas</a>, <a href="/search/cs?searchtype=author&query=de+palma%2C+N">Noel de palma</a>, <a href="/search/cs?searchtype=author&query=Hagimont%2C+D">Daniel Hagimont</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="2104.04060v1-abstract-short" style="display: inline;"> Nowadays, datacenters lean on a computer-centric approach based on monolithic servers which include all necessary hardware resources (mainly CPU, RAM, network and disks) to run applications. Such an architecture comes with two main limitations: (1) difficulty to achieve full resource utilization and (2) coarse granularity for hardware maintenance. Recently, many works investigated a resource-centr… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2104.04060v1-abstract-full').style.display = 'inline'; document.getElementById('2104.04060v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2104.04060v1-abstract-full" style="display: none;"> Nowadays, datacenters lean on a computer-centric approach based on monolithic servers which include all necessary hardware resources (mainly CPU, RAM, network and disks) to run applications. Such an architecture comes with two main limitations: (1) difficulty to achieve full resource utilization and (2) coarse granularity for hardware maintenance. Recently, many works investigated a resource-centric approach called disaggregated architecture where the datacenter is composed of self-content resource boards interconnected using fast interconnection technologies, each resource board including instances of one resource type. The resource-centric architecture allows each resource to be managed (maintenance, allocation) independently. LegoOS is the first work which studied the implications of disaggregation on the operating system, proposing to disaggregate the operating system itself. They demonstrated the suitability of this approach, considering mainly CPU and RAM resources. However, they didn't study the implication of disaggregation on network resources. We reproduced a LegoOS infrastructure and extended it to support disaggregated networking. We show that networking can be disaggregated following the same principles, and that classical networking optimizations such as DMA, DDIO or loopback can be reproduced in such an environment. Our evaluations show the viability of the approach and the potential of future disaggregated infrastructures. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2104.04060v1-abstract-full').style.display = 'none'; document.getElementById('2104.04060v1-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, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 2021. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">10 pages, 8 figures</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2012.00740">arXiv:2012.00740</a> <span> [<a href="https://arxiv.org/pdf/2012.00740">pdf</a>, <a href="https://arxiv.org/format/2012.00740">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Cryptography and Security">cs.CR</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="Machine Learning">cs.LG</span> </div> </div> <p class="title is-5 mathjax"> MYSTIKO : : Cloud-Mediated, Private, Federated Gradient Descent </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Jayaram%2C+K+R">K. R. Jayaram</a>, <a href="/search/cs?searchtype=author&query=Verma%2C+A">Archit Verma</a>, <a href="/search/cs?searchtype=author&query=Verma%2C+A">Ashish Verma</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Gegi Thomas</a>, <a href="/search/cs?searchtype=author&query=Sutcher-Shepard%2C+C">Colin Sutcher-Shepard</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2012.00740v1-abstract-short" style="display: inline;"> Federated learning enables multiple, distributed participants (potentially on different clouds) to collaborate and train machine/deep learning models by sharing parameters/gradients. However, sharing gradients, instead of centralizing data, may not be as private as one would expect. Reverse engineering attacks on plaintext gradients have been demonstrated to be practically feasible. Existing solut… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2012.00740v1-abstract-full').style.display = 'inline'; document.getElementById('2012.00740v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2012.00740v1-abstract-full" style="display: none;"> Federated learning enables multiple, distributed participants (potentially on different clouds) to collaborate and train machine/deep learning models by sharing parameters/gradients. However, sharing gradients, instead of centralizing data, may not be as private as one would expect. Reverse engineering attacks on plaintext gradients have been demonstrated to be practically feasible. Existing solutions for differentially private federated learning, while promising, lead to less accurate models and require nontrivial hyperparameter tuning. In this paper, we examine the use of additive homomorphic encryption (specifically the Paillier cipher) to design secure federated gradient descent techniques that (i) do not require addition of statistical noise or hyperparameter tuning, (ii) does not alter the final accuracy or utility of the final model, (iii) ensure that the plaintext model parameters/gradients of a participant are never revealed to any other participant or third party coordinator involved in the federated learning job, (iv) minimize the trust placed in any third party coordinator and (v) are efficient, with minimal overhead, and cost effective. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2012.00740v1-abstract-full').style.display = 'none'; document.getElementById('2012.00740v1-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 December, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2020. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">IEEE CLOUD 2020</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2009.12446">arXiv:2009.12446</a> <span> [<a href="https://arxiv.org/pdf/2009.12446">pdf</a>, <a href="https://arxiv.org/format/2009.12446">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> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Systems and Control">eess.SY</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/TNSRE.2020.3027501">10.1109/TNSRE.2020.3027501 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> A Complex Stiffness Human Impedance Model with Customizable Exoskeleton Control </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=He%2C+B">Binghan He</a>, <a href="/search/cs?searchtype=author&query=Huang%2C+H">Huang Huang</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G+C">Gray C. Thomas</a>, <a href="/search/cs?searchtype=author&query=Sentis%2C+L">Luis Sentis</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2009.12446v1-abstract-short" style="display: inline;"> The natural impedance, or dynamic relationship between force and motion, of a human operator can determine the stability of exoskeletons that use interaction-torque feedback to amplify human strength. While human impedance is typically modelled as a linear system, our experiments on a single-joint exoskeleton testbed involving 10 human subjects show evidence of nonlinear behavior: a low-frequency… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2009.12446v1-abstract-full').style.display = 'inline'; document.getElementById('2009.12446v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2009.12446v1-abstract-full" style="display: none;"> The natural impedance, or dynamic relationship between force and motion, of a human operator can determine the stability of exoskeletons that use interaction-torque feedback to amplify human strength. While human impedance is typically modelled as a linear system, our experiments on a single-joint exoskeleton testbed involving 10 human subjects show evidence of nonlinear behavior: a low-frequency asymptotic phase for the dynamic stiffness of the human that is different than the expected zero, and an unexpectedly consistent damping ratio as the stiffness and inertia vary. To explain these observations, this paper considers a new frequency-domain model of the human joint dynamics featuring complex value stiffness comprising a real stiffness term and a hysteretic damping term. Using a statistical F-test we show that the hysteretic damping term is not only significant but is even more significant than the linear damping term. Further analysis reveals a linear trend linking hysteretic damping and the real part of the stiffness, which allows us to simplify the complex stiffness model down to a 1-parameter system. Then, we introduce and demonstrate a customizable fractional-order controller that exploits this hysteretic damping behavior to improve strength amplification bandwidth while maintaining stability, and explore a tuning approach which ensures that this stability property is robust to muscle co-contraction for each individual. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2009.12446v1-abstract-full').style.display = 'none'; document.getElementById('2009.12446v1-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 September, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 2020. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">10 pages, 7 figures, 4 tables. arXiv admin note: text overlap with arXiv:1903.00704</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2009.09241">arXiv:2009.09241</a> <span> [<a href="https://arxiv.org/pdf/2009.09241">pdf</a>, <a href="https://arxiv.org/format/2009.09241">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> </div> </div> <p class="title is-5 mathjax"> Word class flexibility: A deep contextualized approach </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Li%2C+B">Bai Li</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Guillaume Thomas</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+Y">Yang Xu</a>, <a href="/search/cs?searchtype=author&query=Rudzicz%2C+F">Frank Rudzicz</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2009.09241v1-abstract-short" style="display: inline;"> Word class flexibility refers to the phenomenon whereby a single word form is used across different grammatical categories. Extensive work in linguistic typology has sought to characterize word class flexibility across languages, but quantifying this phenomenon accurately and at scale has been fraught with difficulties. We propose a principled methodology to explore regularity in word class flexib… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2009.09241v1-abstract-full').style.display = 'inline'; document.getElementById('2009.09241v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2009.09241v1-abstract-full" style="display: none;"> Word class flexibility refers to the phenomenon whereby a single word form is used across different grammatical categories. Extensive work in linguistic typology has sought to characterize word class flexibility across languages, but quantifying this phenomenon accurately and at scale has been fraught with difficulties. We propose a principled methodology to explore regularity in word class flexibility. Our method builds on recent work in contextualized word embeddings to quantify semantic shift between word classes (e.g., noun-to-verb, verb-to-noun), and we apply this method to 37 languages. We find that contextualized embeddings not only capture human judgment of class variation within words in English, but also uncover shared tendencies in class flexibility across languages. Specifically, we find greater semantic variation when flexible lemmas are used in their dominant word class, supporting the view that word class flexibility is a directional process. Our work highlights the utility of deep contextualized models in linguistic typology. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2009.09241v1-abstract-full').style.display = 'none'; document.getElementById('2009.09241v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 19 September, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 2020. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">To appear in EMNLP 2020 (Long Paper)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2007.10987">arXiv:2007.10987</a> <span> [<a href="https://arxiv.org/pdf/2007.10987">pdf</a>, <a href="https://arxiv.org/format/2007.10987">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="Cryptography and Security">cs.CR</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"> IBM Federated Learning: an Enterprise Framework White Paper V0.1 </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Ludwig%2C+H">Heiko Ludwig</a>, <a href="/search/cs?searchtype=author&query=Baracaldo%2C+N">Nathalie Baracaldo</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Gegi Thomas</a>, <a href="/search/cs?searchtype=author&query=Zhou%2C+Y">Yi Zhou</a>, <a href="/search/cs?searchtype=author&query=Anwar%2C+A">Ali Anwar</a>, <a href="/search/cs?searchtype=author&query=Rajamoni%2C+S">Shashank Rajamoni</a>, <a href="/search/cs?searchtype=author&query=Ong%2C+Y">Yuya Ong</a>, <a href="/search/cs?searchtype=author&query=Radhakrishnan%2C+J">Jayaram Radhakrishnan</a>, <a href="/search/cs?searchtype=author&query=Verma%2C+A">Ashish Verma</a>, <a href="/search/cs?searchtype=author&query=Sinn%2C+M">Mathieu Sinn</a>, <a href="/search/cs?searchtype=author&query=Purcell%2C+M">Mark Purcell</a>, <a href="/search/cs?searchtype=author&query=Rawat%2C+A">Ambrish Rawat</a>, <a href="/search/cs?searchtype=author&query=Minh%2C+T">Tran Minh</a>, <a href="/search/cs?searchtype=author&query=Holohan%2C+N">Naoise Holohan</a>, <a href="/search/cs?searchtype=author&query=Chakraborty%2C+S">Supriyo Chakraborty</a>, <a href="/search/cs?searchtype=author&query=Whitherspoon%2C+S">Shalisha Whitherspoon</a>, <a href="/search/cs?searchtype=author&query=Steuer%2C+D">Dean Steuer</a>, <a href="/search/cs?searchtype=author&query=Wynter%2C+L">Laura Wynter</a>, <a href="/search/cs?searchtype=author&query=Hassan%2C+H">Hifaz Hassan</a>, <a href="/search/cs?searchtype=author&query=Laguna%2C+S">Sean Laguna</a>, <a href="/search/cs?searchtype=author&query=Yurochkin%2C+M">Mikhail Yurochkin</a>, <a href="/search/cs?searchtype=author&query=Agarwal%2C+M">Mayank Agarwal</a>, <a href="/search/cs?searchtype=author&query=Chuba%2C+E">Ebube Chuba</a>, <a href="/search/cs?searchtype=author&query=Abay%2C+A">Annie Abay</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2007.10987v1-abstract-short" style="display: inline;"> Federated Learning (FL) is an approach to conduct machine learning without centralizing training data in a single place, for reasons of privacy, confidentiality or data volume. However, solving federated machine learning problems raises issues above and beyond those of centralized machine learning. These issues include setting up communication infrastructure between parties, coordinating the learn… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2007.10987v1-abstract-full').style.display = 'inline'; document.getElementById('2007.10987v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2007.10987v1-abstract-full" style="display: none;"> Federated Learning (FL) is an approach to conduct machine learning without centralizing training data in a single place, for reasons of privacy, confidentiality or data volume. However, solving federated machine learning problems raises issues above and beyond those of centralized machine learning. These issues include setting up communication infrastructure between parties, coordinating the learning process, integrating party results, understanding the characteristics of the training data sets of different participating parties, handling data heterogeneity, and operating with the absence of a verification data set. IBM Federated Learning provides infrastructure and coordination for federated learning. Data scientists can design and run federated learning jobs based on existing, centralized machine learning models and can provide high-level instructions on how to run the federation. The framework applies to both Deep Neural Networks as well as ``traditional'' approaches for the most common machine learning libraries. {\proj} enables data scientists to expand their scope from centralized to federated machine learning, minimizing the learning curve at the outset while also providing the flexibility to deploy to different compute environments and design custom fusion algorithms. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2007.10987v1-abstract-full').style.display = 'none'; document.getElementById('2007.10987v1-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 July, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2020. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">17 pages</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">ACM Class:</span> I.2.6; 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/2006.08875">arXiv:2006.08875</a> <span> [<a href="https://arxiv.org/pdf/2006.08875">pdf</a>, <a href="https://arxiv.org/format/2006.08875">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">stat.ML</span> </div> </div> <p class="title is-5 mathjax"> Model-based Adversarial Meta-Reinforcement Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lin%2C+Z">Zichuan Lin</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Garrett Thomas</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+G">Guangwen Yang</a>, <a href="/search/cs?searchtype=author&query=Ma%2C+T">Tengyu Ma</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="2006.08875v2-abstract-short" style="display: inline;"> Meta-reinforcement learning (meta-RL) aims to learn from multiple training tasks the ability to adapt efficiently to unseen test tasks. Despite the success, existing meta-RL algorithms are known to be sensitive to the task distribution shift. When the test task distribution is different from the training task distribution, the performance may degrade significantly. To address this issue, this pape… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2006.08875v2-abstract-full').style.display = 'inline'; document.getElementById('2006.08875v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2006.08875v2-abstract-full" style="display: none;"> Meta-reinforcement learning (meta-RL) aims to learn from multiple training tasks the ability to adapt efficiently to unseen test tasks. Despite the success, existing meta-RL algorithms are known to be sensitive to the task distribution shift. When the test task distribution is different from the training task distribution, the performance may degrade significantly. To address this issue, this paper proposes Model-based Adversarial Meta-Reinforcement Learning (AdMRL), where we aim to minimize the worst-case sub-optimality gap -- the difference between the optimal return and the return that the algorithm achieves after adaptation -- across all tasks in a family of tasks, with a model-based approach. We propose a minimax objective and optimize it by alternating between learning the dynamics model on a fixed task and finding the adversarial task for the current model -- the task for which the policy induced by the model is maximally suboptimal. Assuming the family of tasks is parameterized, we derive a formula for the gradient of the suboptimality with respect to the task parameters via the implicit function theorem, and show how the gradient estimator can be efficiently implemented by the conjugate gradient method and a novel use of the REINFORCE estimator. We evaluate our approach on several continuous control benchmarks and demonstrate its efficacy in the worst-case performance over all tasks, the generalization power to out-of-distribution tasks, and in training and test time sample efficiency, over existing state-of-the-art meta-RL algorithms. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2006.08875v2-abstract-full').style.display = 'none'; document.getElementById('2006.08875v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 27 February, 2021; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 15 June, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2020. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Accepted by NeurIPS 2020. Code at https://github.com/LinZichuan/AdMRL</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2005.13239">arXiv:2005.13239</a> <span> [<a href="https://arxiv.org/pdf/2005.13239">pdf</a>, <a href="https://arxiv.org/format/2005.13239">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">stat.ML</span> </div> </div> <p class="title is-5 mathjax"> MOPO: Model-based Offline Policy Optimization </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Yu%2C+T">Tianhe Yu</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Garrett Thomas</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+L">Lantao Yu</a>, <a href="/search/cs?searchtype=author&query=Ermon%2C+S">Stefano Ermon</a>, <a href="/search/cs?searchtype=author&query=Zou%2C+J">James Zou</a>, <a href="/search/cs?searchtype=author&query=Levine%2C+S">Sergey Levine</a>, <a href="/search/cs?searchtype=author&query=Finn%2C+C">Chelsea Finn</a>, <a href="/search/cs?searchtype=author&query=Ma%2C+T">Tengyu Ma</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2005.13239v6-abstract-short" style="display: inline;"> Offline reinforcement learning (RL) refers to the problem of learning policies entirely from a large batch of previously collected data. This problem setting offers the promise of utilizing such datasets to acquire policies without any costly or dangerous active exploration. However, it is also challenging, due to the distributional shift between the offline training data and those states visited… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2005.13239v6-abstract-full').style.display = 'inline'; document.getElementById('2005.13239v6-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2005.13239v6-abstract-full" style="display: none;"> Offline reinforcement learning (RL) refers to the problem of learning policies entirely from a large batch of previously collected data. This problem setting offers the promise of utilizing such datasets to acquire policies without any costly or dangerous active exploration. However, it is also challenging, due to the distributional shift between the offline training data and those states visited by the learned policy. Despite significant recent progress, the most successful prior methods are model-free and constrain the policy to the support of data, precluding generalization to unseen states. In this paper, we first observe that an existing model-based RL algorithm already produces significant gains in the offline setting compared to model-free approaches. However, standard model-based RL methods, designed for the online setting, do not provide an explicit mechanism to avoid the offline setting's distributional shift issue. Instead, we propose to modify the existing model-based RL methods by applying them with rewards artificially penalized by the uncertainty of the dynamics. We theoretically show that the algorithm maximizes a lower bound of the policy's return under the true MDP. We also characterize the trade-off between the gain and risk of leaving the support of the batch data. Our algorithm, Model-based Offline Policy Optimization (MOPO), outperforms standard model-based RL algorithms and prior state-of-the-art model-free offline RL algorithms on existing offline RL benchmarks and two challenging continuous control tasks that require generalizing from data collected for a different task. The code is available at https://github.com/tianheyu927/mopo. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2005.13239v6-abstract-full').style.display = 'none'; document.getElementById('2005.13239v6-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 November, 2020; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 27 May, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 2020. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">NeurIPS 2020. First two authors contributed equally. Last two authors advised equally</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1907.04964">arXiv:1907.04964</a> <span> [<a href="https://arxiv.org/pdf/1907.04964">pdf</a>, <a href="https://arxiv.org/format/1907.04964">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">stat.ML</span> </div> </div> <p class="title is-5 mathjax"> A Model-based Approach for Sample-efficient Multi-task Reinforcement Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Landolfi%2C+N+C">Nicholas C. Landolfi</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Garrett Thomas</a>, <a href="/search/cs?searchtype=author&query=Ma%2C+T">Tengyu Ma</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="1907.04964v3-abstract-short" style="display: inline;"> The aim of multi-task reinforcement learning is two-fold: (1) efficiently learn by training against multiple tasks and (2) quickly adapt, using limited samples, to a variety of new tasks. In this work, the tasks correspond to reward functions for environments with the same (or similar) dynamical models. We propose to learn a dynamical model during the training process and use this model to perform… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1907.04964v3-abstract-full').style.display = 'inline'; document.getElementById('1907.04964v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1907.04964v3-abstract-full" style="display: none;"> The aim of multi-task reinforcement learning is two-fold: (1) efficiently learn by training against multiple tasks and (2) quickly adapt, using limited samples, to a variety of new tasks. In this work, the tasks correspond to reward functions for environments with the same (or similar) dynamical models. We propose to learn a dynamical model during the training process and use this model to perform sample-efficient adaptation to new tasks at test time. We use significantly fewer samples by performing policy optimization only in a "virtual" environment whose transitions are given by our learned dynamical model. Our algorithm sequentially trains against several tasks. Upon encountering a new task, we first warm-up a policy on our learned dynamical model, which requires no new samples from the environment. We then adapt the dynamical model with samples from this policy in the real environment. We evaluate our approach on several continuous control benchmarks and demonstrate its efficacy over MAML, a state-of-the-art meta-learning algorithm, on these tasks. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1907.04964v3-abstract-full').style.display = 'none'; document.getElementById('1907.04964v3-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 3 November, 2019; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 10 July, 2019; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2019. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">13 pages, 3 figures</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1903.09673">arXiv:1903.09673</a> <span> [<a href="https://arxiv.org/pdf/1903.09673">pdf</a>, <a href="https://arxiv.org/format/1903.09673">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> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Systems and Control">eess.SY</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Dynamical Systems">math.DS</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Optimization and Control">math.OC</span> </div> </div> <p class="title is-5 mathjax"> Compliance Shaping for Control of Strength Amplification Exoskeletons with Elastic Cuffs </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Thomas%2C+G+C">Gray Cortright Thomas</a>, <a href="/search/cs?searchtype=author&query=Coholich%2C+J+M">Jeremiah M. Coholich</a>, <a href="/search/cs?searchtype=author&query=Sentis%2C+L">Luis Sentis</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="1903.09673v1-abstract-short" style="display: inline;"> Exoskeletons which amplify the strength of their operators can enable heavy-duty manipulation of unknown objects. However, this type of behavior is difficult to accomplish; it requires the exoskeleton to sense and amplify the operator's interaction forces while remaining stable. But, the goals of amplification and robust stability when connected to the operator fundamentally conflict. As a solutio… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1903.09673v1-abstract-full').style.display = 'inline'; document.getElementById('1903.09673v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1903.09673v1-abstract-full" style="display: none;"> Exoskeletons which amplify the strength of their operators can enable heavy-duty manipulation of unknown objects. However, this type of behavior is difficult to accomplish; it requires the exoskeleton to sense and amplify the operator's interaction forces while remaining stable. But, the goals of amplification and robust stability when connected to the operator fundamentally conflict. As a solution, we introduce a design with a spring in series with the force sensitive cuff. This allows us to design an exoskeleton compliance behavior which is nominally passive, even with high amplification ratios. In practice, time delay and discrete time filters prevent our strategy from actually achieving passivity, but the designed compliance still makes the exoskeleton more robust to spring-like human behaviors. Our exoskeleton is actuated by a series elastic actuator (SEA), which introduces another spring into the system. We show that shaping the cuff compliance for the exoskeleton can be made into approximately the same problem as shaping the spring compliance of an SEA. We therefore introduce a feedback controller and gain tuning method which takes advantage of an existing compliance shaping technique for SEAs. We call our strategy the "double compliance shaping" method. With large amplification ratios, this controller tends to amplify nonlinear transmission friction effects, so we additionally propose a "transmission disturbance observer" to mitigate this drawback. Our methods are validated on a single-degree-of-freedom elbow exoskeleton. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1903.09673v1-abstract-full').style.display = 'none'; document.getElementById('1903.09673v1-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 March, 2019; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 2019. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">8 pages, 9 figures, conference</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">MSC Class:</span> 70Q05; 70E60; 93C80 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1903.00704">arXiv:1903.00704</a> <span> [<a href="https://arxiv.org/pdf/1903.00704">pdf</a>, <a href="https://arxiv.org/format/1903.00704">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 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/IROS40897.2019.8968005">10.1109/IROS40897.2019.8968005 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Complex Stiffness Model of Physical Human-Robot Interaction: Implications for Control of Performance Augmentation Exoskeletons </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=He%2C+B">Binghan He</a>, <a href="/search/cs?searchtype=author&query=Huang%2C+H">Huang Huang</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G+C">Gray C. Thomas</a>, <a href="/search/cs?searchtype=author&query=Sentis%2C+L">Luis Sentis</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="1903.00704v4-abstract-short" style="display: inline;"> Human joint dynamic stiffness plays an important role in the stability of performance augmentation exoskeletons. In this paper, we consider a new frequency domain model of the human joint dynamics which features a complex value stiffness. This complex stiffness consists of a real stiffness and a hysteretic damping. We use it to explain the dynamic behaviors of the human connected to the exoskeleto… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1903.00704v4-abstract-full').style.display = 'inline'; document.getElementById('1903.00704v4-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1903.00704v4-abstract-full" style="display: none;"> Human joint dynamic stiffness plays an important role in the stability of performance augmentation exoskeletons. In this paper, we consider a new frequency domain model of the human joint dynamics which features a complex value stiffness. This complex stiffness consists of a real stiffness and a hysteretic damping. We use it to explain the dynamic behaviors of the human connected to the exoskeleton, in particular the observed non-zero low frequency phase shift and the near constant damping ratio of the resonant as stiffness and inertia vary. We validate this concept by experimenting with an elbow-joint exoskeleton testbed on a subject while modifying joint stiffness behavior, exoskeleton inertia, and strength augmentation gains. We compare three different models of elbow-joint dynamic stiffness: a model with real stiffness, viscous damping and inertia, a model with complex stiffness and inertia, and a model combining the previous two models. Our results show that the hysteretic damping term improves modeling accuracy, using a statistical F-test. Moreover this improvement is statistically more significant than using classical viscous damping term. In addition, we experimentally observe a linear relationship between the hysteretic damping and the real part of the stiffness which allows us to simplify the complex stiffness model as a 1-parameter system. Ultimately, we design a fractional order controller to demonstrate how human hysteretic damping behavior can be exploited to improve strength amplification performance while maintaining stability. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1903.00704v4-abstract-full').style.display = 'none'; document.getElementById('1903.00704v4-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, 2020; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 2 March, 2019; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 2019. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Accepted for publication in IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS) Copyright 2019 IEEE</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1901.06261">arXiv:1901.06261</a> <span> [<a href="https://arxiv.org/pdf/1901.06261">pdf</a>, <a href="https://arxiv.org/format/1901.06261">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="Software Engineering">cs.SE</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"> NeuNetS: An Automated Synthesis Engine for Neural Network Design </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Sood%2C+A">Atin Sood</a>, <a href="/search/cs?searchtype=author&query=Elder%2C+B">Benjamin Elder</a>, <a href="/search/cs?searchtype=author&query=Herta%2C+B">Benjamin Herta</a>, <a href="/search/cs?searchtype=author&query=Xue%2C+C">Chao Xue</a>, <a href="/search/cs?searchtype=author&query=Bekas%2C+C">Costas Bekas</a>, <a href="/search/cs?searchtype=author&query=Malossi%2C+A+C+I">A. Cristiano I. Malossi</a>, <a href="/search/cs?searchtype=author&query=Saha%2C+D">Debashish Saha</a>, <a href="/search/cs?searchtype=author&query=Scheidegger%2C+F">Florian Scheidegger</a>, <a href="/search/cs?searchtype=author&query=Venkataraman%2C+G">Ganesh Venkataraman</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Gegi Thomas</a>, <a href="/search/cs?searchtype=author&query=Mariani%2C+G">Giovanni Mariani</a>, <a href="/search/cs?searchtype=author&query=Strobelt%2C+H">Hendrik Strobelt</a>, <a href="/search/cs?searchtype=author&query=Samulowitz%2C+H">Horst Samulowitz</a>, <a href="/search/cs?searchtype=author&query=Wistuba%2C+M">Martin Wistuba</a>, <a href="/search/cs?searchtype=author&query=Manica%2C+M">Matteo Manica</a>, <a href="/search/cs?searchtype=author&query=Choudhury%2C+M">Mihir Choudhury</a>, <a href="/search/cs?searchtype=author&query=Yan%2C+R">Rong Yan</a>, <a href="/search/cs?searchtype=author&query=Istrate%2C+R">Roxana Istrate</a>, <a href="/search/cs?searchtype=author&query=Puri%2C+R">Ruchir Puri</a>, <a href="/search/cs?searchtype=author&query=Pedapati%2C+T">Tejaswini Pedapati</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="1901.06261v1-abstract-short" style="display: inline;"> Application of neural networks to a vast variety of practical applications is transforming the way AI is applied in practice. Pre-trained neural network models available through APIs or capability to custom train pre-built neural network architectures with customer data has made the consumption of AI by developers much simpler and resulted in broad adoption of these complex AI models. While prebui… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1901.06261v1-abstract-full').style.display = 'inline'; document.getElementById('1901.06261v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1901.06261v1-abstract-full" style="display: none;"> Application of neural networks to a vast variety of practical applications is transforming the way AI is applied in practice. Pre-trained neural network models available through APIs or capability to custom train pre-built neural network architectures with customer data has made the consumption of AI by developers much simpler and resulted in broad adoption of these complex AI models. While prebuilt network models exist for certain scenarios, to try and meet the constraints that are unique to each application, AI teams need to think about developing custom neural network architectures that can meet the tradeoff between accuracy and memory footprint to achieve the tight constraints of their unique use-cases. However, only a small proportion of data science teams have the skills and experience needed to create a neural network from scratch, and the demand far exceeds the supply. In this paper, we present NeuNetS : An automated Neural Network Synthesis engine for custom neural network design that is available as part of IBM's AI OpenScale's product. NeuNetS is available for both Text and Image domains and can build neural networks for specific tasks in a fraction of the time it takes today with human effort, and with accuracy similar to that of human-designed AI models. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1901.06261v1-abstract-full').style.display = 'none'; document.getElementById('1901.06261v1-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 January, 2019; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> January 2019. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">14 pages, 12 figures. arXiv admin note: text overlap with arXiv:1806.00250</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1812.01719">arXiv:1812.01719</a> <span> [<a href="https://arxiv.org/pdf/1812.01719">pdf</a>, <a href="https://arxiv.org/format/1812.01719">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">stat.ML</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/10.3389/fninf.2019.00067">10.3389/fninf.2019.00067 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Knowing what you know in brain segmentation using Bayesian deep neural networks </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=McClure%2C+P">Patrick McClure</a>, <a href="/search/cs?searchtype=author&query=Rho%2C+N">Nao Rho</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+J+A">John A. Lee</a>, <a href="/search/cs?searchtype=author&query=Kaczmarzyk%2C+J+R">Jakub R. Kaczmarzyk</a>, <a href="/search/cs?searchtype=author&query=Zheng%2C+C">Charles Zheng</a>, <a href="/search/cs?searchtype=author&query=Ghosh%2C+S+S">Satrajit S. Ghosh</a>, <a href="/search/cs?searchtype=author&query=Nielson%2C+D">Dylan Nielson</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+A+G">Adam G. Thomas</a>, <a href="/search/cs?searchtype=author&query=Bandettini%2C+P">Peter Bandettini</a>, <a href="/search/cs?searchtype=author&query=Pereira%2C+F">Francisco Pereira</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="1812.01719v5-abstract-short" style="display: inline;"> In this paper, we describe a Bayesian deep neural network (DNN) for predicting FreeSurfer segmentations of structural MRI volumes, in minutes rather than hours. The network was trained and evaluated on a large dataset (n = 11,480), obtained by combining data from more than a hundred different sites, and also evaluated on another completely held-out dataset (n = 418). The network was trained using… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1812.01719v5-abstract-full').style.display = 'inline'; document.getElementById('1812.01719v5-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1812.01719v5-abstract-full" style="display: none;"> In this paper, we describe a Bayesian deep neural network (DNN) for predicting FreeSurfer segmentations of structural MRI volumes, in minutes rather than hours. The network was trained and evaluated on a large dataset (n = 11,480), obtained by combining data from more than a hundred different sites, and also evaluated on another completely held-out dataset (n = 418). The network was trained using a novel spike-and-slab dropout-based variational inference approach. We show that, on these datasets, the proposed Bayesian DNN outperforms previously proposed methods, in terms of the similarity between the segmentation predictions and the FreeSurfer labels, and the usefulness of the estimate uncertainty of these predictions. In particular, we demonstrated that the prediction uncertainty of this network at each voxel is a good indicator of whether the network has made an error and that the uncertainty across the whole brain can predict the manual quality control ratings of a scan. The proposed Bayesian DNN method should be applicable to any new network architecture for addressing the segmentation problem. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1812.01719v5-abstract-full').style.display = 'none'; document.getElementById('1812.01719v5-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 18 September, 2019; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 3 December, 2018; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2018. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Submitted to Frontiers in Neuroinformatics</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1803.07635">arXiv:1803.07635</a> <span> [<a href="https://arxiv.org/pdf/1803.07635">pdf</a>, <a href="https://arxiv.org/format/1803.07635">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> <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"> Learning Robotic Assembly from CAD </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Garrett Thomas</a>, <a href="/search/cs?searchtype=author&query=Chien%2C+M">Melissa Chien</a>, <a href="/search/cs?searchtype=author&query=Tamar%2C+A">Aviv Tamar</a>, <a href="/search/cs?searchtype=author&query=Ojea%2C+J+A">Juan Aparicio Ojea</a>, <a href="/search/cs?searchtype=author&query=Abbeel%2C+P">Pieter Abbeel</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="1803.07635v2-abstract-short" style="display: inline;"> In this work, motivated by recent manufacturing trends, we investigate autonomous robotic assembly. Industrial assembly tasks require contact-rich manipulation skills, which are challenging to acquire using classical control and motion planning approaches. Consequently, robot controllers for assembly domains are presently engineered to solve a particular task, and cannot easily handle variations i… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1803.07635v2-abstract-full').style.display = 'inline'; document.getElementById('1803.07635v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1803.07635v2-abstract-full" style="display: none;"> In this work, motivated by recent manufacturing trends, we investigate autonomous robotic assembly. Industrial assembly tasks require contact-rich manipulation skills, which are challenging to acquire using classical control and motion planning approaches. Consequently, robot controllers for assembly domains are presently engineered to solve a particular task, and cannot easily handle variations in the product or environment. Reinforcement learning (RL) is a promising approach for autonomously acquiring robot skills that involve contact-rich dynamics. However, RL relies on random exploration for learning a control policy, which requires many robot executions, and often gets trapped in locally suboptimal solutions. Instead, we posit that prior knowledge, when available, can improve RL performance. We exploit the fact that in modern assembly domains, geometric information about the task is readily available via the CAD design files. We propose to leverage this prior knowledge by guiding RL along a geometric motion plan, calculated using the CAD data. We show that our approach effectively improves over traditional control approaches for tracking the motion plan, and can solve assembly tasks that require high precision, even without accurate state estimation. In addition, we propose a neural network architecture that can learn to track the motion plan, and generalize the assembly controller to changes in the object positions. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1803.07635v2-abstract-full').style.display = 'none'; document.getElementById('1803.07635v2-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, 2018; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 20 March, 2018; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> March 2018. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">In the proceedings of the IEEE International Conference on Robotics and Automation (ICRA), Brisbane, Australia, May 2018</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1802.10190">arXiv:1802.10190</a> <span> [<a href="https://arxiv.org/pdf/1802.10190">pdf</a>, <a href="https://arxiv.org/format/1802.10190">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> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Optimization and Control">math.OC</span> </div> </div> <p class="title is-5 mathjax"> Exploiting the Natural Dynamics of Series Elastic Robots by Actuator-Centered Sequential Linear Programming </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Schlossman%2C+R">Rachel Schlossman</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G+C">Gray C. Thomas</a>, <a href="/search/cs?searchtype=author&query=Campbell%2C+O">Orion Campbell</a>, <a href="/search/cs?searchtype=author&query=Sentis%2C+L">Luis Sentis</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="1802.10190v2-abstract-short" style="display: inline;"> Series elastic robots are best able to follow trajectories which obey the limitations of their actuators, since they cannot instantly change their joint forces. In fact, the performance of series elastic actuators can surpass that of ideal force source actuators by storing and releasing energy. In this paper, we formulate the trajectory optimization problem for series elastic robots in a novel way… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1802.10190v2-abstract-full').style.display = 'inline'; document.getElementById('1802.10190v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1802.10190v2-abstract-full" style="display: none;"> Series elastic robots are best able to follow trajectories which obey the limitations of their actuators, since they cannot instantly change their joint forces. In fact, the performance of series elastic actuators can surpass that of ideal force source actuators by storing and releasing energy. In this paper, we formulate the trajectory optimization problem for series elastic robots in a novel way based on sequential linear programming. Our framework is unique in the separation of the actuator dynamics from the rest of the dynamics, and in the use of a tunable pseudo-mass parameter that improves the discretization accuracy of our approach. The actuator dynamics are truly linear, which allows them to be excluded from trust-region mechanics. This causes our algorithm to have similar run times with and without the actuator dynamics. We demonstrate our optimization algorithm by tuning high performance behaviors for a single-leg robot in simulation and on hardware for a single degree-of-freedom actuator testbed. The results show that compliance allows for faster motions and takes a similar amount of computation time. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1802.10190v2-abstract-full').style.display = 'none'; document.getElementById('1802.10190v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 16 July, 2018; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 27 February, 2018; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2018. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1712.04989">arXiv:1712.04989</a> <span> [<a href="https://arxiv.org/pdf/1712.04989">pdf</a>, <a href="https://arxiv.org/format/1712.04989">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"> Persistent Memory Programming Abstractions in Context of Concurrent Applications </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Singh%2C+A">Ajay Singh</a>, <a href="/search/cs?searchtype=author&query=Shapiro%2C+M">Marc Shapiro</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Gael Thomas</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="1712.04989v1-abstract-short" style="display: inline;"> The advent of non-volatile memory (NVM) technologies like PCM, STT, memristors and Fe-RAM is believed to enhance the system performance by getting rid of the traditional memory hierarchy by reducing the gap between memory and storage. This memory technology is considered to have the performance like that of DRAM and persistence like that of disks. Thus, it would also provide significant performanc… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1712.04989v1-abstract-full').style.display = 'inline'; document.getElementById('1712.04989v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1712.04989v1-abstract-full" style="display: none;"> The advent of non-volatile memory (NVM) technologies like PCM, STT, memristors and Fe-RAM is believed to enhance the system performance by getting rid of the traditional memory hierarchy by reducing the gap between memory and storage. This memory technology is considered to have the performance like that of DRAM and persistence like that of disks. Thus, it would also provide significant performance benefits for big data applications by allowing in-memory processing of large data with the lowest latency to persistence. Leveraging the performance benefits of this memory-centric computing technology through traditional memory programming is not trivial and the challenges aggravate for parallel/concurrent applications. To this end, several programming abstractions have been proposed like NVthreads, Mnemosyne and intel's NVML. However, deciding upon a programming abstraction which is easier to program and at the same time ensures the consistency and balances various software and architectural trade-offs is openly debatable and active area of research for NVM community. We study the NVthreads, Mnemosyne and NVML libraries by building a concurrent and persistent set and open addressed hash-table data structure application. In this process, we explore and report various tradeoffs and hidden costs involved in building concurrent applications for persistence in terms of achieving efficiency, consistency and ease of programming with these NVM programming abstractions. Eventually, we evaluate the performance of the set and hash-table data structure applications. We observe that NVML is easiest to program with but is least efficient and Mnemosyne is most performance friendly but involves significant programming efforts to build concurrent and persistent applications. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1712.04989v1-abstract-full').style.display = 'none'; document.getElementById('1712.04989v1-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 December, 2017; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2017. </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 in HiPC SRS 2017</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1609.09001">arXiv:1609.09001</a> <span> [<a href="https://arxiv.org/pdf/1609.09001">pdf</a>, <a href="https://arxiv.org/format/1609.09001">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> <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"> Learning from the Hindsight Plan -- Episodic MPC Improvement </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Tamar%2C+A">Aviv Tamar</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Garrett Thomas</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+T">Tianhao Zhang</a>, <a href="/search/cs?searchtype=author&query=Levine%2C+S">Sergey Levine</a>, <a href="/search/cs?searchtype=author&query=Abbeel%2C+P">Pieter Abbeel</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="1609.09001v2-abstract-short" style="display: inline;"> Model predictive control (MPC) is a popular control method that has proved effective for robotics, among other fields. MPC performs re-planning at every time step. Re-planning is done with a limited horizon per computational and real-time constraints and often also for robustness to potential model errors. However, the limited horizon leads to suboptimal performance. In this work, we consider the… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1609.09001v2-abstract-full').style.display = 'inline'; document.getElementById('1609.09001v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1609.09001v2-abstract-full" style="display: none;"> Model predictive control (MPC) is a popular control method that has proved effective for robotics, among other fields. MPC performs re-planning at every time step. Re-planning is done with a limited horizon per computational and real-time constraints and often also for robustness to potential model errors. However, the limited horizon leads to suboptimal performance. In this work, we consider the iterative learning setting, where the same task can be repeated several times, and propose a policy improvement scheme for MPC. The main idea is that between executions we can, offline, run MPC with a longer horizon, resulting in a hindsight plan. To bring the next real-world execution closer to the hindsight plan, our approach learns to re-shape the original cost function with the goal of satisfying the following property: short horizon planning (as realistic during real executions) with respect to the shaped cost should result in mimicking the hindsight plan. This effectively consolidates long-term reasoning into the short-horizon planning. We empirically evaluate our approach in contact-rich manipulation tasks both in simulated and real environments, such as peg insertion by a real PR2 robot. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1609.09001v2-abstract-full').style.display = 'none'; document.getElementById('1609.09001v2-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 March, 2017; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 28 September, 2016; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 2016. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Additional experiments for neural network generalization and for varying the planning horizon. Paper accepted to ICRA 2017</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1602.02867">arXiv:1602.02867</a> <span> [<a href="https://arxiv.org/pdf/1602.02867">pdf</a>, <a href="https://arxiv.org/format/1602.02867">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> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Neural and Evolutionary Computing">cs.NE</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">stat.ML</span> </div> </div> <p class="title is-5 mathjax"> Value Iteration Networks </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Tamar%2C+A">Aviv Tamar</a>, <a href="/search/cs?searchtype=author&query=Wu%2C+Y">Yi Wu</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Garrett Thomas</a>, <a href="/search/cs?searchtype=author&query=Levine%2C+S">Sergey Levine</a>, <a href="/search/cs?searchtype=author&query=Abbeel%2C+P">Pieter Abbeel</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="1602.02867v4-abstract-short" style="display: inline;"> We introduce the value iteration network (VIN): a fully differentiable neural network with a `planning module' embedded within. VINs can learn to plan, and are suitable for predicting outcomes that involve planning-based reasoning, such as policies for reinforcement learning. Key to our approach is a novel differentiable approximation of the value-iteration algorithm, which can be represented as a… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1602.02867v4-abstract-full').style.display = 'inline'; document.getElementById('1602.02867v4-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1602.02867v4-abstract-full" style="display: none;"> We introduce the value iteration network (VIN): a fully differentiable neural network with a `planning module' embedded within. VINs can learn to plan, and are suitable for predicting outcomes that involve planning-based reasoning, such as policies for reinforcement learning. Key to our approach is a novel differentiable approximation of the value-iteration algorithm, which can be represented as a convolutional neural network, and trained end-to-end using standard backpropagation. We evaluate VIN based policies on discrete and continuous path-planning domains, and on a natural-language based search task. We show that by learning an explicit planning computation, VIN policies generalize better to new, unseen domains. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1602.02867v4-abstract-full').style.display = 'none'; document.getElementById('1602.02867v4-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 March, 2017; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 9 February, 2016; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2016. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Fixed missing table values</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> Advances in Neural Information Processing Systems 29 pages 2154--2162, 2016 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1501.02855">arXiv:1501.02855</a> <span> [<a href="https://arxiv.org/pdf/1501.02855">pdf</a>, <a href="https://arxiv.org/format/1501.02855">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> <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"> Assessing Whole-Body Operational Space Control in a Point-Foot Series Elastic Biped: Balance on Split Terrain and Undirected Walking </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Kim%2C+D">Donghyun Kim</a>, <a href="/search/cs?searchtype=author&query=Zhao%2C+Y">Ye Zhao</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Gray Thomas</a>, <a href="/search/cs?searchtype=author&query=Sentis%2C+L">Luis Sentis</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="1501.02855v1-abstract-short" style="display: inline;"> In this paper we present advancements in control and trajectory generation for agile behavior in bipedal robots. We demonstrate that Whole-Body Operational Space Control (WBOSC), developed a few years ago, is well suited for achieving two types of agile behaviors, namely, balancing on a high pitch split terrain and achieving undirected walking on flat terrain. The work presented here is the first… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1501.02855v1-abstract-full').style.display = 'inline'; document.getElementById('1501.02855v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1501.02855v1-abstract-full" style="display: none;"> In this paper we present advancements in control and trajectory generation for agile behavior in bipedal robots. We demonstrate that Whole-Body Operational Space Control (WBOSC), developed a few years ago, is well suited for achieving two types of agile behaviors, namely, balancing on a high pitch split terrain and achieving undirected walking on flat terrain. The work presented here is the first implementation of WBOSC on a biped robot, and more specifically a biped robot with series elastic actuators. We present and analyze a new algorithm that dynamically balances point foot robots by choosing footstep placements. Dealing with the naturally unstable dynamics of these type of systems is a difficult problem that requires both the controller and the trajectory generation algorithm to operate quickly and efficiently. We put forth a comprehensive development and integration effort: the design and construction of the biped system and experimental infrastructure, a customization of WBOSC for the agile behaviors, and new trajectory generation algorithms. Using this custom built controller, we conduct, for first time, an experiment in which a biped robot balances in a high pitch split terrain, demonstrating our ability to precisely regulate internal forces using force sensing feedback techniques. Finally, we demonstrate the stabilizing capabilities of our online trajectory generation algorithm in the physics-based simulator and through physical experiments with a planarized locomotion setup. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1501.02855v1-abstract-full').style.display = 'none'; document.getElementById('1501.02855v1-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 January, 2015; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> January 2015. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">17 pages, 9 figures, 4 tables</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1407.4346">arXiv:1407.4346</a> <span> [<a href="https://arxiv.org/pdf/1407.4346">pdf</a>, <a href="https://arxiv.org/ps/1407.4346">ps</a>, <a href="https://arxiv.org/format/1407.4346">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Software Engineering">cs.SE</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Operating Systems">cs.OS</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/10.1145/2619090">10.1145/2619090 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Faults in Linux 2.6 </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Palix%2C+N">Nicolas Palix</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Ga毛l Thomas</a>, <a href="/search/cs?searchtype=author&query=Saha%2C+S">Suman Saha</a>, <a href="/search/cs?searchtype=author&query=Calv%C3%A8s%2C+C">Christophe Calv猫s</a>, <a href="/search/cs?searchtype=author&query=Muller%2C+G">Gilles Muller</a>, <a href="/search/cs?searchtype=author&query=Lawall%2C+J+L">Julia L. Lawall</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="1407.4346v1-abstract-short" style="display: inline;"> In August 2011, Linux entered its third decade. Ten years before, Chou et al. published a study of faults found by applying a static analyzer to Linux versions 1.0 through 2.4.1. A major result of their work was that the drivers directory contained up to 7 times more of certain kinds of faults than other directories. This result inspired numerous efforts on improving the reliability of driver code… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1407.4346v1-abstract-full').style.display = 'inline'; document.getElementById('1407.4346v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1407.4346v1-abstract-full" style="display: none;"> In August 2011, Linux entered its third decade. Ten years before, Chou et al. published a study of faults found by applying a static analyzer to Linux versions 1.0 through 2.4.1. A major result of their work was that the drivers directory contained up to 7 times more of certain kinds of faults than other directories. This result inspired numerous efforts on improving the reliability of driver code. Today, Linux is used in a wider range of environments, provides a wider range of services, and has adopted a new development and release model. What has been the impact of these changes on code quality? To answer this question, we have transported Chou et al.'s experiments to all versions of Linux 2.6; released between 2003 and 2011. We find that Linux has more than doubled in size during this period, but the number of faults per line of code has been decreasing. Moreover, the fault rate of drivers is now below that of other directories, such as arch. These results can guide further development and research efforts for the decade to come. To allow updating these results as Linux evolves, we define our experimental protocol and make our checkers available. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1407.4346v1-abstract-full').style.display = 'none'; document.getElementById('1407.4346v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 16 July, 2014; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2014. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> ACM Transactions on Computer Systems 32, 2 (2014) 1--40 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1310.8392">arXiv:1310.8392</a> <span> [<a href="https://arxiv.org/pdf/1310.8392">pdf</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="Cryptography and Security">cs.CR</span> </div> </div> <p class="title is-5 mathjax"> Cloud computing security using encryption technique </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Geethu Thomas</a>, <a href="/search/cs?searchtype=author&query=Jose%2C+P">Prem Jose V</a>, <a href="/search/cs?searchtype=author&query=Afsar%2C+P">P. Afsar</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="1310.8392v1-abstract-short" style="display: inline;"> Cloud Computing has been envisioned as the next generation architecture of IT Enterprise. The Cloud computing concept offers dynamically scalable resources provisioned as a service over the Internet. Economic benefits are the main driver for the Cloud, since it promises the reduction of capital expenditure and operational expenditure. In order for this to become reality, however, there are still s… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1310.8392v1-abstract-full').style.display = 'inline'; document.getElementById('1310.8392v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1310.8392v1-abstract-full" style="display: none;"> Cloud Computing has been envisioned as the next generation architecture of IT Enterprise. The Cloud computing concept offers dynamically scalable resources provisioned as a service over the Internet. Economic benefits are the main driver for the Cloud, since it promises the reduction of capital expenditure and operational expenditure. In order for this to become reality, however, there are still some challenges to be solved. Most important among these are security and trust issues,since the users data has to be released to the Cloud and thus leaves the protection sphere of the data owner.In contrast to traditional solutions, where the IT services are under proper physical,logical and personnel controls, Cloud Computing moves the application software and databases to the large data centers, where the management of the data and services may not be fully trustworthy. This unique attribute, however, poses many new security challenges which have not been well understood. Security is to save data from danger and vulnerability. There are so many dangers and vulnerabilities to be handled. Various security issues and some of their solution are explained and are concentrating mainly on public cloud security issues and their solutions. Data should always be encrypted when stored(using separate symmetric encryption keys)and transmitted. If this is implemented appropriately, even if another tenant can access the data, all that will appear is gibberish. So a method is proposed such that we are encrypting the whole data along with the cryptographic key. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1310.8392v1-abstract-full').style.display = 'none'; document.getElementById('1310.8392v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 31 October, 2013; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2013. </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, 3 Figures. arXiv admin note: text overlap with arXiv:1303.4814 by other authors</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/0904.4058">arXiv:0904.4058</a> <span> [<a href="https://arxiv.org/pdf/0904.4058">pdf</a>, <a href="https://arxiv.org/format/0904.4058">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Cryptography and Security">cs.CR</span> </div> </div> <p class="title is-5 mathjax"> Security impact ratings considered harmful </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Arnold%2C+J">Jeff Arnold</a>, <a href="/search/cs?searchtype=author&query=Abbott%2C+T">Tim Abbott</a>, <a href="/search/cs?searchtype=author&query=Daher%2C+W">Waseem Daher</a>, <a href="/search/cs?searchtype=author&query=Price%2C+G">Gregory Price</a>, <a href="/search/cs?searchtype=author&query=Elhage%2C+N">Nelson Elhage</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Geoffrey Thomas</a>, <a href="/search/cs?searchtype=author&query=Kaseorg%2C+A">Anders Kaseorg</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="0904.4058v1-abstract-short" style="display: inline;"> In this paper, we question the common practice of assigning security impact ratings to OS updates. Specifically, we present evidence that ranking updates by their perceived security importance, in order to defer applying some updates, exposes systems to significant risk. We argue that OS vendors and security groups should not focus on security updates to the detriment of other updates, but sho… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('0904.4058v1-abstract-full').style.display = 'inline'; document.getElementById('0904.4058v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="0904.4058v1-abstract-full" style="display: none;"> In this paper, we question the common practice of assigning security impact ratings to OS updates. Specifically, we present evidence that ranking updates by their perceived security importance, in order to defer applying some updates, exposes systems to significant risk. We argue that OS vendors and security groups should not focus on security updates to the detriment of other updates, but should instead seek update technologies that make it feasible to distribute updates for all disclosed OS bugs in a timely manner. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('0904.4058v1-abstract-full').style.display = 'none'; document.getElementById('0904.4058v1-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 April, 2009; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 2009. </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">HotOS 2009</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/0802.3475">arXiv:0802.3475</a> <span> [<a href="https://arxiv.org/pdf/0802.3475">pdf</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Software Engineering">cs.SE</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Human-Computer Interaction">cs.HC</span> </div> </div> <p class="title is-5 mathjax"> Spreadsheet Development Methodologies using Resolver: Moving spreadsheets into the 21st Century </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Kemmis%2C+P">Patrick Kemmis</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Giles Thomas</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="0802.3475v1-abstract-short" style="display: inline;"> We intend to demonstrate the innate problems with existing spreadsheet products and to show how to tackle these issues using a new type of spreadsheet program called Resolver. It addresses the issues head-on and thereby moves the 1980's "VisiCalc paradigm" on to match the advances in computer languages and user requirements. Continuous display of the spreadsheet grid and the equivalent computer… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('0802.3475v1-abstract-full').style.display = 'inline'; document.getElementById('0802.3475v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="0802.3475v1-abstract-full" style="display: none;"> We intend to demonstrate the innate problems with existing spreadsheet products and to show how to tackle these issues using a new type of spreadsheet program called Resolver. It addresses the issues head-on and thereby moves the 1980's "VisiCalc paradigm" on to match the advances in computer languages and user requirements. Continuous display of the spreadsheet grid and the equivalent computer program, together with the ability to interact and add code through either interface, provides a number of new methodologies for spreadsheet development. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('0802.3475v1-abstract-full').style.display = 'none'; document.getElementById('0802.3475v1-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 February, 2008; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2008. </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</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">ACM Class:</span> J.1; H.4.1; K.6.4; D.2.9 </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> Proc. European Spreadsheet Risks Int. Grp. (EuSpRIG) 2007 93-104 ISBN 978-905617-58-6 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/cs/0411081">arXiv:cs/0411081</a> <span> [<a href="https://arxiv.org/pdf/cs/0411081">pdf</a>, <a href="https://arxiv.org/ps/cs/0411081">ps</a>, <a href="https://arxiv.org/format/cs/0411081">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Networking and Internet Architecture">cs.NI</span> </div> </div> <p class="title is-5 mathjax"> Reconfigurations dynamiques de services dans un intergiciel a composants CORBA CCM </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Hachichi%2C+A">Assia Hachichi</a>, <a href="/search/cs?searchtype=author&query=Martin%2C+C">Cyril Martin</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+G">Gael Thomas</a>, <a href="/search/cs?searchtype=author&query=Patarin%2C+S">Simon Patarin</a>, <a href="/search/cs?searchtype=author&query=Folliot%2C+B">Bertil Folliot</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="cs/0411081v1-abstract-short" style="display: inline;"> Today, component oriented middlewares are used to design, develop and deploy easily distributed applications, by ensuring the heterogeneity, interoperability, and reuse of the software modules, and the separation between the business code encapsulated in the components and the system code managed by the containers. Several standards answer this definition such as: CCM (CORBA Component Model), EJ… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('cs/0411081v1-abstract-full').style.display = 'inline'; document.getElementById('cs/0411081v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="cs/0411081v1-abstract-full" style="display: none;"> Today, component oriented middlewares are used to design, develop and deploy easily distributed applications, by ensuring the heterogeneity, interoperability, and reuse of the software modules, and the separation between the business code encapsulated in the components and the system code managed by the containers. Several standards answer this definition such as: CCM (CORBA Component Model), EJB (Enterprise Java Beans) and .Net. However these standards offer a limited and fixed number of system services, removing any possibility to add system services or to reconfigure dynamically the middleware. Our works propose mechanisms to add and to adapt dynamically the system services, based on a reconfiguration language which is dynamically adaptable to the need of the reconfiguration, and on a tool of dynamic reconfiguration, a prototype was achieved for the OpenCCM platform, that is an implementation of the CCM specification. This work was partially financed by the european project IST-COACH (2001-34445). <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('cs/0411081v1-abstract-full').style.display = 'none'; document.getElementById('cs/0411081v1-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 November, 2004; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2004. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> DECOR04 (2004) 159-170 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/cmp-lg/9801001">arXiv:cmp-lg/9801001</a> <span> [<a href="https://arxiv.org/pdf/cmp-lg/9801001">pdf</a>, <a href="https://arxiv.org/ps/cmp-lg/9801001">ps</a>, <a href="https://arxiv.org/format/cmp-lg/9801001">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> </div> </div> <p class="title is-5 mathjax"> Hierarchical Non-Emitting Markov Models </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Ristad%2C+E+S">Eric Sven Ristad</a>, <a href="/search/cs?searchtype=author&query=Thomas%2C+R+G">Robert G. Thomas</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="cmp-lg/9801001v3-abstract-short" style="display: inline;"> We describe a simple variant of the interpolated Markov model with non-emitting state transitions and prove that it is strictly more powerful than any Markov model. More importantly, the non-emitting model outperforms the classic interpolated model on the natural language texts under a wide range of experimental conditions, with only a modest increase in computational requirements. The non-emitt… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('cmp-lg/9801001v3-abstract-full').style.display = 'inline'; document.getElementById('cmp-lg/9801001v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="cmp-lg/9801001v3-abstract-full" style="display: none;"> We describe a simple variant of the interpolated Markov model with non-emitting state transitions and prove that it is strictly more powerful than any Markov model. More importantly, the non-emitting model outperforms the classic interpolated model on the natural language texts under a wide range of experimental conditions, with only a modest increase in computational requirements. The non-emitting model is also much less prone to overfitting. Keywords: Markov model, interpolated Markov model, hidden Markov model, mixture modeling, non-emitting state transitions, state-conditional interpolation, statistical language model, discrete time series, Brown corpus, Wall Street Journal. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('cmp-lg/9801001v3-abstract-full').style.display = 'none'; document.getElementById('cmp-lg/9801001v3-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 January, 1998; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 14 January, 1998; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> January 1998. </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">http://www.cs.princeton.edu/~ristad/papers/pu-544-97.ps.gz</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Report number:</span> CS-TR-544-97 </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=Thomas%2C+G&start=50" class="pagination-next" >Next </a> <ul class="pagination-list"> <li> <a href="/search/?searchtype=author&query=Thomas%2C+G&start=0" class="pagination-link is-current" aria-label="Goto page 1">1 </a> </li> <li> <a href="/search/?searchtype=author&query=Thomas%2C+G&start=50" class="pagination-link " aria-label="Page 2" aria-current="page">2 </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