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 1–39 of 39 results for author: <span class="mathjax">Chiang, H</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=Chiang%2C+H">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="Chiang, H"> </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=Chiang%2C+H&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="Chiang, H"> <ul id="abstracts"><li><input checked id="abstracts-0" name="abstracts" type="radio" value="show"> <label for="abstracts-0">Show abstracts</label></li><li><input id="abstracts-1" name="abstracts" type="radio" value="hide"> <label for="abstracts-1">Hide abstracts</label></li></ul> </div> <div class="box field is-grouped is-grouped-multiline level-item"> <div class="control"> <span class="select is-small"> <select id="size" name="size"><option value="25">25</option><option selected value="50">50</option><option value="100">100</option><option value="200">200</option></select> </span> <label for="size">results per page</label>. </div> <div class="control"> <label for="order">Sort results by</label> <span class="select is-small"> <select id="order" name="order"><option selected value="-announced_date_first">Announcement date (newest first)</option><option value="announced_date_first">Announcement date (oldest first)</option><option value="-submitted_date">Submission date (newest first)</option><option value="submitted_date">Submission date (oldest first)</option><option value="">Relevance</option></select> </span> </div> <div class="control"> <button class="button is-small is-link">Go</button> </div> </div> </form> </div> </div> <ol class="breathe-horizontal" start="1"> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2410.13229">arXiv:2410.13229</a> <span> [<a href="https://arxiv.org/pdf/2410.13229">pdf</a>, <a href="https://arxiv.org/format/2410.13229">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> Quamba: A Post-Training Quantization Recipe for Selective State Space Models </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Hung-Yueh Chiang</a>, <a href="/search/cs?searchtype=author&query=Chang%2C+C">Chi-Chih Chang</a>, <a href="/search/cs?searchtype=author&query=Frumkin%2C+N">Natalia Frumkin</a>, <a href="/search/cs?searchtype=author&query=Wu%2C+K">Kai-Chiang Wu</a>, <a href="/search/cs?searchtype=author&query=Marculescu%2C+D">Diana Marculescu</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2410.13229v1-abstract-short" style="display: inline;"> State Space Models (SSMs) have emerged as an appealing alternative to Transformers for large language models, achieving state-of-the-art accuracy with constant memory complexity which allows for holding longer context lengths than attention-based networks. The superior computational efficiency of SSMs in long sequence modeling positions them favorably over Transformers in many scenarios. However,… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.13229v1-abstract-full').style.display = 'inline'; document.getElementById('2410.13229v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2410.13229v1-abstract-full" style="display: none;"> State Space Models (SSMs) have emerged as an appealing alternative to Transformers for large language models, achieving state-of-the-art accuracy with constant memory complexity which allows for holding longer context lengths than attention-based networks. The superior computational efficiency of SSMs in long sequence modeling positions them favorably over Transformers in many scenarios. However, improving the efficiency of SSMs on request-intensive cloud-serving and resource-limited edge applications is still a formidable task. SSM quantization is a possible solution to this problem, making SSMs more suitable for wide deployment, while still maintaining their accuracy. Quantization is a common technique to reduce the model size and to utilize the low bit-width acceleration features on modern computing units, yet existing quantization techniques are poorly suited for SSMs. Most notably, SSMs have highly sensitive feature maps within the selective scan mechanism (i.e., linear recurrence) and massive outliers in the output activations which are not present in the output of token-mixing in the self-attention modules. To address this issue, we propose a static 8-bit per-tensor SSM quantization method which suppresses the maximum values of the input activations to the selective SSM for finer quantization precision and quantizes the output activations in an outlier-free space with Hadamard transform. Our 8-bit weight-activation quantized Mamba 2.8B SSM benefits from hardware acceleration and achieves a 1.72x lower generation latency on an Nvidia Orin Nano 8G, with only a 0.9% drop in average accuracy on zero-shot tasks. The experiments demonstrate the effectiveness and practical applicability of our approach for deploying SSM-based models of all sizes on both cloud and edge platforms. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.13229v1-abstract-full').style.display = 'none'; document.getElementById('2410.13229v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 17 October, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2410.01150">arXiv:2410.01150</a> <span> [<a href="https://arxiv.org/pdf/2410.01150">pdf</a>, <a href="https://arxiv.org/format/2410.01150">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Audio and Speech Processing">eess.AS</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Sound">cs.SD</span> </div> </div> <p class="title is-5 mathjax"> Restorative Speech Enhancement: A Progressive Approach Using SE and Codec Modules </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Hsin-Tien Chiang</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+H">Hao Zhang</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+Y">Yong Xu</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+M">Meng Yu</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+D">Dong Yu</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2410.01150v1-abstract-short" style="display: inline;"> In challenging environments with significant noise and reverberation, traditional speech enhancement (SE) methods often lead to over-suppressed speech, creating artifacts during listening and harming downstream tasks performance. To overcome these limitations, we propose a novel approach called Restorative SE (RestSE), which combines a lightweight SE module with a generative codec module to progre… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.01150v1-abstract-full').style.display = 'inline'; document.getElementById('2410.01150v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2410.01150v1-abstract-full" style="display: none;"> In challenging environments with significant noise and reverberation, traditional speech enhancement (SE) methods often lead to over-suppressed speech, creating artifacts during listening and harming downstream tasks performance. To overcome these limitations, we propose a novel approach called Restorative SE (RestSE), which combines a lightweight SE module with a generative codec module to progressively enhance and restore speech quality. The SE module initially reduces noise, while the codec module subsequently performs dereverberation and restores speech using generative capabilities. We systematically explore various quantization techniques within the codec module to optimize performance. Additionally, we introduce a weighted loss function and feature fusion that merges the SE output with the original mixture, particularly at segments where the SE output is heavily distorted. Experimental results demonstrate the effectiveness of our proposed method in enhancing speech quality under adverse conditions. Audio demos are available at: https://sophie091524.github.io/RestorativeSE/. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2410.01150v1-abstract-full').style.display = 'none'; document.getElementById('2410.01150v1-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 October, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2024. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Paper in submission</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2409.07832">arXiv:2409.07832</a> <span> [<a href="https://arxiv.org/pdf/2409.07832">pdf</a>, <a href="https://arxiv.org/format/2409.07832">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Hardware Architecture">cs.AR</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> </div> </div> <p class="title is-5 mathjax"> Efficient and Reliable Vector Similarity Search Using Asymmetric Encoding with NAND-Flash for Many-Class Few-Shot Learning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Hao-Wei Chiang</a>, <a href="/search/cs?searchtype=author&query=Huang%2C+C">Chi-Tse Huang</a>, <a href="/search/cs?searchtype=author&query=Cheng%2C+H">Hsiang-Yun Cheng</a>, <a href="/search/cs?searchtype=author&query=Tseng%2C+P">Po-Hao Tseng</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+M">Ming-Hsiu Lee</a>, <a href="/search/cs?searchtype=author&query=An-Yeu"> An-Yeu</a>, <a href="/search/cs?searchtype=author&query=Wu"> Wu</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2409.07832v1-abstract-short" style="display: inline;"> While memory-augmented neural networks (MANNs) offer an effective solution for few-shot learning (FSL) by integrating deep neural networks with external memory, the capacity requirements and energy overhead of data movement become enormous due to the large number of support vectors in many-class FSL scenarios. Various in-memory search solutions have emerged to improve the energy efficiency of MANN… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2409.07832v1-abstract-full').style.display = 'inline'; document.getElementById('2409.07832v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2409.07832v1-abstract-full" style="display: none;"> While memory-augmented neural networks (MANNs) offer an effective solution for few-shot learning (FSL) by integrating deep neural networks with external memory, the capacity requirements and energy overhead of data movement become enormous due to the large number of support vectors in many-class FSL scenarios. Various in-memory search solutions have emerged to improve the energy efficiency of MANNs. NAND-based multi-bit content addressable memory (MCAM) is a promising option due to its high density and large capacity. Despite its potential, MCAM faces limitations such as a restricted number of word lines, limited quantization levels, and non-ideal effects like varying string currents and bottleneck effects, which lead to significant accuracy drops. To address these issues, we propose several innovative methods. First, the Multi-bit Thermometer Code (MTMC) leverages the extensive capacity of MCAM to enhance vector precision using cumulative encoding rules, thereby mitigating the bottleneck effect. Second, the Asymmetric vector similarity search (AVSS) reduces the precision of the query vector while maintaining that of the support vectors, thereby minimizing the search iterations and improving efficiency in many-class scenarios. Finally, the Hardware-Aware Training (HAT) method optimizes controller training by modeling the hardware characteristics of MCAM, thus enhancing the reliability of the system. Our integrated framework reduces search iterations by up to 32 times, and increases overall accuracy by 1.58% to 6.94%. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2409.07832v1-abstract-full').style.display = 'none'; document.getElementById('2409.07832v1-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 September, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2408.15395">arXiv:2408.15395</a> <span> [<a href="https://arxiv.org/pdf/2408.15395">pdf</a>, <a href="https://arxiv.org/format/2408.15395">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> SCAN-Edge: Finding MobileNet-speed Hybrid Networks for Diverse Edge Devices via Hardware-Aware Evolutionary Search </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Hung-Yueh Chiang</a>, <a href="/search/cs?searchtype=author&query=Marculescu%2C+D">Diana Marculescu</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2408.15395v1-abstract-short" style="display: inline;"> Designing low-latency and high-efficiency hybrid networks for a variety of low-cost commodity edge devices is both costly and tedious, leading to the adoption of hardware-aware neural architecture search (NAS) for finding optimal architectures. However, unifying NAS for a wide range of edge devices presents challenges due to the variety of hardware designs, supported operations, and compilation op… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2408.15395v1-abstract-full').style.display = 'inline'; document.getElementById('2408.15395v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2408.15395v1-abstract-full" style="display: none;"> Designing low-latency and high-efficiency hybrid networks for a variety of low-cost commodity edge devices is both costly and tedious, leading to the adoption of hardware-aware neural architecture search (NAS) for finding optimal architectures. However, unifying NAS for a wide range of edge devices presents challenges due to the variety of hardware designs, supported operations, and compilation optimizations. Existing methods often fix the search space of architecture choices (e.g., activation, convolution, or self-attention) and estimate latency using hardware-agnostic proxies (e.g., FLOPs), which fail to achieve proclaimed latency across various edge devices. To address this issue, we propose SCAN-Edge, a unified NAS framework that jointly searches for self-attention, convolution, and activation to accommodate the wide variety of edge devices, including CPU-, GPU-, and hardware accelerator-based systems. To handle the large search space, SCAN-Edge relies on with a hardware-aware evolutionary algorithm that improves the quality of the search space to accelerate the sampling process. Experiments on large-scale datasets demonstrate that our hybrid networks match the actual MobileNetV2 latency for 224x224 input resolution on various commodity edge devices. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2408.15395v1-abstract-full').style.display = 'none'; document.getElementById('2408.15395v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 27 August, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> August 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2407.07775">arXiv:2407.07775</a> <span> [<a href="https://arxiv.org/pdf/2407.07775">pdf</a>, <a href="https://arxiv.org/format/2407.07775">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> </div> </div> <p class="title is-5 mathjax"> Mobility VLA: Multimodal Instruction Navigation with Long-Context VLMs and Topological Graphs </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chiang%2C+H+L">Hao-Tien Lewis Chiang</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+Z">Zhuo Xu</a>, <a href="/search/cs?searchtype=author&query=Fu%2C+Z">Zipeng Fu</a>, <a href="/search/cs?searchtype=author&query=Jacob%2C+M+G">Mithun George Jacob</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+T">Tingnan Zhang</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+T+E">Tsang-Wei Edward Lee</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+W">Wenhao Yu</a>, <a href="/search/cs?searchtype=author&query=Schenck%2C+C">Connor Schenck</a>, <a href="/search/cs?searchtype=author&query=Rendleman%2C+D">David Rendleman</a>, <a href="/search/cs?searchtype=author&query=Shah%2C+D">Dhruv Shah</a>, <a href="/search/cs?searchtype=author&query=Xia%2C+F">Fei Xia</a>, <a href="/search/cs?searchtype=author&query=Hsu%2C+J">Jasmine Hsu</a>, <a href="/search/cs?searchtype=author&query=Hoech%2C+J">Jonathan Hoech</a>, <a href="/search/cs?searchtype=author&query=Florence%2C+P">Pete Florence</a>, <a href="/search/cs?searchtype=author&query=Kirmani%2C+S">Sean Kirmani</a>, <a href="/search/cs?searchtype=author&query=Singh%2C+S">Sumeet Singh</a>, <a href="/search/cs?searchtype=author&query=Sindhwani%2C+V">Vikas Sindhwani</a>, <a href="/search/cs?searchtype=author&query=Parada%2C+C">Carolina Parada</a>, <a href="/search/cs?searchtype=author&query=Finn%2C+C">Chelsea Finn</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+P">Peng Xu</a>, <a href="/search/cs?searchtype=author&query=Levine%2C+S">Sergey Levine</a>, <a href="/search/cs?searchtype=author&query=Tan%2C+J">Jie Tan</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2407.07775v2-abstract-short" style="display: inline;"> An elusive goal in navigation research is to build an intelligent agent that can understand multimodal instructions including natural language and image, and perform useful navigation. To achieve this, we study a widely useful category of navigation tasks we call Multimodal Instruction Navigation with demonstration Tours (MINT), in which the environment prior is provided through a previously recor… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2407.07775v2-abstract-full').style.display = 'inline'; document.getElementById('2407.07775v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2407.07775v2-abstract-full" style="display: none;"> An elusive goal in navigation research is to build an intelligent agent that can understand multimodal instructions including natural language and image, and perform useful navigation. To achieve this, we study a widely useful category of navigation tasks we call Multimodal Instruction Navigation with demonstration Tours (MINT), in which the environment prior is provided through a previously recorded demonstration video. Recent advances in Vision Language Models (VLMs) have shown a promising path in achieving this goal as it demonstrates capabilities in perceiving and reasoning about multimodal inputs. However, VLMs are typically trained to predict textual output and it is an open research question about how to best utilize them in navigation. To solve MINT, we present Mobility VLA, a hierarchical Vision-Language-Action (VLA) navigation policy that combines the environment understanding and common sense reasoning power of long-context VLMs and a robust low-level navigation policy based on topological graphs. The high-level policy consists of a long-context VLM that takes the demonstration tour video and the multimodal user instruction as input to find the goal frame in the tour video. Next, a low-level policy uses the goal frame and an offline constructed topological graph to generate robot actions at every timestep. We evaluated Mobility VLA in a 836m^2 real world environment and show that Mobility VLA has a high end-to-end success rates on previously unsolved multimodal instructions such as "Where should I return this?" while holding a plastic bin. A video demonstrating Mobility VLA can be found here: https://youtu.be/-Tof__Q8_5s <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2407.07775v2-abstract-full').style.display = 'none'; document.getElementById('2407.07775v2-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 July, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 10 July, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> July 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2405.15773">arXiv:2405.15773</a> <span> [<a href="https://arxiv.org/pdf/2405.15773">pdf</a>, <a href="https://arxiv.org/format/2405.15773">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"> Feature Aggregation with Latent Generative Replay for Federated Continual Learning of Socially Appropriate Robot Behaviours </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Churamani%2C+N">Nikhil Churamani</a>, <a href="/search/cs?searchtype=author&query=Checker%2C+S">Saksham Checker</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+H+L">Hao-Tien Lewis Chiang</a>, <a href="/search/cs?searchtype=author&query=Gunes%2C+H">Hatice Gunes</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="2405.15773v1-abstract-short" style="display: inline;"> For widespread real-world applications, it is beneficial for robots to explore Federated Learning (FL) settings where several robots, deployed in parallel, can learn independently while also sharing their learning with each other. This work explores a simulated living room environment where robots need to learn the social appropriateness of their actions. We propose Federated Root (FedRoot), a nov… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2405.15773v1-abstract-full').style.display = 'inline'; document.getElementById('2405.15773v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2405.15773v1-abstract-full" style="display: none;"> For widespread real-world applications, it is beneficial for robots to explore Federated Learning (FL) settings where several robots, deployed in parallel, can learn independently while also sharing their learning with each other. This work explores a simulated living room environment where robots need to learn the social appropriateness of their actions. We propose Federated Root (FedRoot), a novel weight aggregation strategy which disentangles feature learning across clients from individual task-based learning. Adapting popular FL strategies to use FedRoot instead, we present a novel FL benchmark for learning the social appropriateness of different robot actions in diverse social configurations. FedRoot-based methods offer competitive performance compared to others while offering sizeable (up to 86% for CPU usage and up to 72% for GPU usage) reduction in resource consumption. Furthermore, real-world interactions require social robots to dynamically adapt to changing environmental and task settings. To facilitate this, we propose Federated Latent Generative Replay (FedLGR), a novel Federated Continual Learning (FCL) strategy that uses FedRoot-based weight aggregation and embeds each client with a generator model for pseudo-rehearsal of learnt feature embeddings to mitigate forgetting in a resource-efficient manner. Our benchmark results demonstrate that FedRoot-based FCL methods outperform other methods while also offering sizeable (up to 84% for CPU usage and up to 92% for GPU usage) reduction in resource consumption, with FedLGR providing the best results across evaluations. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2405.15773v1-abstract-full').style.display = 'none'; document.getElementById('2405.15773v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 16 March, 2024; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 2024. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2311.14683">arXiv:2311.14683</a> <span> [<a href="https://arxiv.org/pdf/2311.14683">pdf</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computers and Society">cs.CY</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Social and Information Networks">cs.SI</span> </div> </div> <p class="title is-5 mathjax"> Data Science for Social Good </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Abbasi%2C+A">Ahmed Abbasi</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+R+H+L">Roger H. L. Chiang</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+J+J">Jennifer J. 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="2311.14683v1-abstract-short" style="display: inline;"> Data science has been described as the fourth paradigm for scientific discovery. The latest wave of data science research, pertaining to machine learning and artificial intelligence (AI), is growing exponentially and garnering millions of annual citations. However, this growth has been accompanied by a diminishing emphasis on social good challenges - our analysis reveals that the proportion of dat… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2311.14683v1-abstract-full').style.display = 'inline'; document.getElementById('2311.14683v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2311.14683v1-abstract-full" style="display: none;"> Data science has been described as the fourth paradigm for scientific discovery. The latest wave of data science research, pertaining to machine learning and artificial intelligence (AI), is growing exponentially and garnering millions of annual citations. However, this growth has been accompanied by a diminishing emphasis on social good challenges - our analysis reveals that the proportion of data science research focusing on social good is less than it has ever been. At the same time, the proliferation of machine learning and generative AI have sparked debates about the socio-technical prospects and challenges associated with data science for human flourishing, organizations, and society. Against this backdrop, we present a framework for "data science for social good" (DSSG) research that considers the interplay between relevant data science research genres, social good challenges, and different levels of socio-technical abstraction. We perform an analysis of the literature to empirically demonstrate the paucity of work on DSSG in information systems (and other related disciplines) and highlight current impediments. We then use our proposed framework to introduce the articles appearing in the special issue. We hope that this article and the special issue will spur future DSSG research and help reverse the alarming trend across data science research over the past 30-plus years in which social good challenges are garnering proportionately less attention with each passing day. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2311.14683v1-abstract-full').style.display = 'none'; document.getElementById('2311.14683v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 2 November, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2311.08878">arXiv:2311.08878</a> <span> [<a href="https://arxiv.org/pdf/2311.08878">pdf</a>, <a href="https://arxiv.org/format/2311.08878">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Audio and Speech Processing">eess.AS</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Sound">cs.SD</span> </div> </div> <p class="title is-5 mathjax"> Multi-objective Non-intrusive Hearing-aid Speech Assessment Model </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Hsin-Tien Chiang</a>, <a href="/search/cs?searchtype=author&query=Fu%2C+S">Szu-Wei Fu</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+H">Hsin-Min Wang</a>, <a href="/search/cs?searchtype=author&query=Tsao%2C+Y">Yu Tsao</a>, <a href="/search/cs?searchtype=author&query=Hansen%2C+J+H+L">John H. L. Hansen</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="2311.08878v1-abstract-short" style="display: inline;"> Without the need for a clean reference, non-intrusive speech assessment methods have caught great attention for objective evaluations. While deep learning models have been used to develop non-intrusive speech assessment methods with promising results, there is limited research on hearing-impaired subjects. This study proposes a multi-objective non-intrusive hearing-aid speech assessment model, cal… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2311.08878v1-abstract-full').style.display = 'inline'; document.getElementById('2311.08878v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2311.08878v1-abstract-full" style="display: none;"> Without the need for a clean reference, non-intrusive speech assessment methods have caught great attention for objective evaluations. While deep learning models have been used to develop non-intrusive speech assessment methods with promising results, there is limited research on hearing-impaired subjects. This study proposes a multi-objective non-intrusive hearing-aid speech assessment model, called HASA-Net Large, which predicts speech quality and intelligibility scores based on input speech signals and specified hearing-loss patterns. Our experiments showed the utilization of pre-trained SSL models leads to a significant boost in speech quality and intelligibility predictions compared to using spectrograms as input. Additionally, we examined three distinct fine-tuning approaches that resulted in further performance improvements. Furthermore, we demonstrated that incorporating SSL models resulted in greater transferability to OOD dataset. Finally, this study introduces HASA-Net Large, which is a non-invasive approach for evaluating speech quality and intelligibility. HASA-Net Large utilizes raw waveforms and hearing-loss patterns to accurately predict speech quality and intelligibility levels for individuals with normal and impaired hearing and demonstrates superior prediction performance and transferability. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2311.08878v1-abstract-full').style.display = 'none'; document.getElementById('2311.08878v1-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 November, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2310.17341">arXiv:2310.17341</a> <span> [<a href="https://arxiv.org/pdf/2310.17341">pdf</a>, <a href="https://arxiv.org/format/2310.17341">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"> De-novo Chemical Reaction Generation by Means of Temporal Convolutional Neural Networks </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Buin%2C+A">Andrei Buin</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+H+Y">Hung Yi Chiang</a>, <a href="/search/cs?searchtype=author&query=Gadsden%2C+S+A">S. Andrew Gadsden</a>, <a href="/search/cs?searchtype=author&query=Alderson%2C+F+A">Faraz A. Alderson</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="2310.17341v3-abstract-short" style="display: inline;"> We present here a combination of two networks, Recurrent Neural Networks (RNN) and Temporarily Convolutional Neural Networks (TCN) in de novo reaction generation using the novel Reaction Smiles-like representation of reactions (CGRSmiles) with atom mapping directly incorporated. Recurrent Neural Networks are known for their autoregressive properties and are frequently used in language modelling wi… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.17341v3-abstract-full').style.display = 'inline'; document.getElementById('2310.17341v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2310.17341v3-abstract-full" style="display: none;"> We present here a combination of two networks, Recurrent Neural Networks (RNN) and Temporarily Convolutional Neural Networks (TCN) in de novo reaction generation using the novel Reaction Smiles-like representation of reactions (CGRSmiles) with atom mapping directly incorporated. Recurrent Neural Networks are known for their autoregressive properties and are frequently used in language modelling with direct application to SMILES generation. The relatively novel TCNs possess similar properties with wide receptive field while obeying the causality required for natural language processing (NLP). The combination of both latent representations expressed through TCN and RNN results in an overall better performance compared to RNN alone. Additionally, it is shown that different fine-tuning protocols have a profound impact on generative scope of the model when applied on a dataset of interest via transfer learning. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.17341v3-abstract-full').style.display = 'none'; document.getElementById('2310.17341v3-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 1 November, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 26 October, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2310.11590">arXiv:2310.11590</a> <span> [<a href="https://arxiv.org/pdf/2310.11590">pdf</a>, <a href="https://arxiv.org/format/2310.11590">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"> Predicting Human Impressions of Robot Performance During Navigation Tasks </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zhang%2C+Q">Qiping Zhang</a>, <a href="/search/cs?searchtype=author&query=Tsoi%2C+N">Nathan Tsoi</a>, <a href="/search/cs?searchtype=author&query=Nagib%2C+M">Mofeed Nagib</a>, <a href="/search/cs?searchtype=author&query=Choi%2C+B">Booyeon Choi</a>, <a href="/search/cs?searchtype=author&query=Tan%2C+J">Jie Tan</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+H+L">Hao-Tien Lewis Chiang</a>, <a href="/search/cs?searchtype=author&query=V%C3%A1zquez%2C+M">Marynel V谩zquez</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="2310.11590v2-abstract-short" style="display: inline;"> Human impressions of robot performance are often measured through surveys. As a more scalable and cost-effective alternative, we investigate the possibility of predicting people's impressions of robot behavior using non-verbal behavioral cues and machine learning techniques. To this end, we first contribute the SEAN TOGETHER Dataset consisting of observations of an interaction between a person and… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.11590v2-abstract-full').style.display = 'inline'; document.getElementById('2310.11590v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2310.11590v2-abstract-full" style="display: none;"> Human impressions of robot performance are often measured through surveys. As a more scalable and cost-effective alternative, we investigate the possibility of predicting people's impressions of robot behavior using non-verbal behavioral cues and machine learning techniques. To this end, we first contribute the SEAN TOGETHER Dataset consisting of observations of an interaction between a person and a mobile robot in a VR simulation, together with impressions of robot performance provided by users on a 5-point scale. Second, we contribute analyses of how well humans and supervised learning techniques can predict perceived robot performance based on different observation types (like facial expression features, and features that describe the navigation behavior of the robot and pedestrians). Our results suggest that facial expressions alone provide useful information about human impressions of robot performance; but in the navigation scenarios that we considered, reasoning about spatial features in context is critical for the prediction task. Also, supervised learning techniques showed promise because they outperformed humans' predictions of robot performance in most cases. Further, when predicting robot performance as a binary classification task on unseen users' data, the F1 Score of machine learning models more than doubled in comparison to predicting performance on a 5-point scale. This suggested that the models can have good generalization capabilities, although they are better at telling the directionality of robot performance than predicting exact performance ratings. Based on our findings in simulation, we conducted a real-world demonstration in which a mobile robot uses a machine learning model to predict how a human that follows it perceives it. Finally, we discuss the implications of our results for implementing such supervised learning models in real-world navigation scenarios. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2310.11590v2-abstract-full').style.display = 'none'; document.getElementById('2310.11590v2-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 November, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 17 October, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2023. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2309.15275">arXiv:2309.15275</a> <span> [<a href="https://arxiv.org/pdf/2309.15275">pdf</a>, <a href="https://arxiv.org/format/2309.15275">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> </div> </div> <p class="title is-5 mathjax"> Efficient Low-rank Backpropagation for Vision Transformer Adaptation </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Yang%2C+Y">Yuedong Yang</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Hung-Yueh Chiang</a>, <a href="/search/cs?searchtype=author&query=Li%2C+G">Guihong Li</a>, <a href="/search/cs?searchtype=author&query=Marculescu%2C+D">Diana Marculescu</a>, <a href="/search/cs?searchtype=author&query=Marculescu%2C+R">Radu Marculescu</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2309.15275v1-abstract-short" style="display: inline;"> The increasing scale of vision transformers (ViT) has made the efficient fine-tuning of these large models for specific needs a significant challenge in various applications. This issue originates from the computationally demanding matrix multiplications required during the backpropagation process through linear layers in ViT. In this paper, we tackle this problem by proposing a new Low-rank BackP… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2309.15275v1-abstract-full').style.display = 'inline'; document.getElementById('2309.15275v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2309.15275v1-abstract-full" style="display: none;"> The increasing scale of vision transformers (ViT) has made the efficient fine-tuning of these large models for specific needs a significant challenge in various applications. This issue originates from the computationally demanding matrix multiplications required during the backpropagation process through linear layers in ViT. In this paper, we tackle this problem by proposing a new Low-rank BackPropagation via Walsh-Hadamard Transformation (LBP-WHT) method. Intuitively, LBP-WHT projects the gradient into a low-rank space and carries out backpropagation. This approach substantially reduces the computation needed for adapting ViT, as matrix multiplication in the low-rank space is far less resource-intensive. We conduct extensive experiments with different models (ViT, hybrid convolution-ViT model) on multiple datasets to demonstrate the effectiveness of our method. For instance, when adapting an EfficientFormer-L1 model on CIFAR100, our LBP-WHT achieves 10.4% higher accuracy than the state-of-the-art baseline, while requiring 9 MFLOPs less computation. As the first work to accelerate ViT adaptation with low-rank backpropagation, our LBP-WHT method is complementary to many prior efforts and can be combined with them for better performance. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2309.15275v1-abstract-full').style.display = 'none'; document.getElementById('2309.15275v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 26 September, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 2023. </p> <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">37th Conference on Neural Information Processing Systems (NeurIPS 2023)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2306.16740">arXiv:2306.16740</a> <span> [<a href="https://arxiv.org/pdf/2306.16740">pdf</a>, <a href="https://arxiv.org/format/2306.16740">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="Human-Computer Interaction">cs.HC</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"> Principles and Guidelines for Evaluating Social Robot Navigation Algorithms </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Francis%2C+A">Anthony Francis</a>, <a href="/search/cs?searchtype=author&query=P%C3%A9rez-D%27Arpino%2C+C">Claudia P茅rez-D'Arpino</a>, <a href="/search/cs?searchtype=author&query=Li%2C+C">Chengshu Li</a>, <a href="/search/cs?searchtype=author&query=Xia%2C+F">Fei Xia</a>, <a href="/search/cs?searchtype=author&query=Alahi%2C+A">Alexandre Alahi</a>, <a href="/search/cs?searchtype=author&query=Alami%2C+R">Rachid Alami</a>, <a href="/search/cs?searchtype=author&query=Bera%2C+A">Aniket Bera</a>, <a href="/search/cs?searchtype=author&query=Biswas%2C+A">Abhijat Biswas</a>, <a href="/search/cs?searchtype=author&query=Biswas%2C+J">Joydeep Biswas</a>, <a href="/search/cs?searchtype=author&query=Chandra%2C+R">Rohan Chandra</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+H+L">Hao-Tien Lewis Chiang</a>, <a href="/search/cs?searchtype=author&query=Everett%2C+M">Michael Everett</a>, <a href="/search/cs?searchtype=author&query=Ha%2C+S">Sehoon Ha</a>, <a href="/search/cs?searchtype=author&query=Hart%2C+J">Justin Hart</a>, <a href="/search/cs?searchtype=author&query=How%2C+J+P">Jonathan P. How</a>, <a href="/search/cs?searchtype=author&query=Karnan%2C+H">Haresh Karnan</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+T+E">Tsang-Wei Edward Lee</a>, <a href="/search/cs?searchtype=author&query=Manso%2C+L+J">Luis J. Manso</a>, <a href="/search/cs?searchtype=author&query=Mirksy%2C+R">Reuth Mirksy</a>, <a href="/search/cs?searchtype=author&query=Pirk%2C+S">S枚ren Pirk</a>, <a href="/search/cs?searchtype=author&query=Singamaneni%2C+P+T">Phani Teja Singamaneni</a>, <a href="/search/cs?searchtype=author&query=Stone%2C+P">Peter Stone</a>, <a href="/search/cs?searchtype=author&query=Taylor%2C+A+V">Ada V. Taylor</a>, <a href="/search/cs?searchtype=author&query=Trautman%2C+P">Peter Trautman</a>, <a href="/search/cs?searchtype=author&query=Tsoi%2C+N">Nathan Tsoi</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="2306.16740v4-abstract-short" style="display: inline;"> A major challenge to deploying robots widely is navigation in human-populated environments, commonly referred to as social robot navigation. While the field of social navigation has advanced tremendously in recent years, the fair evaluation of algorithms that tackle social navigation remains hard because it involves not just robotic agents moving in static environments but also dynamic human agent… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2306.16740v4-abstract-full').style.display = 'inline'; document.getElementById('2306.16740v4-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2306.16740v4-abstract-full" style="display: none;"> A major challenge to deploying robots widely is navigation in human-populated environments, commonly referred to as social robot navigation. While the field of social navigation has advanced tremendously in recent years, the fair evaluation of algorithms that tackle social navigation remains hard because it involves not just robotic agents moving in static environments but also dynamic human agents and their perceptions of the appropriateness of robot behavior. In contrast, clear, repeatable, and accessible benchmarks have accelerated progress in fields like computer vision, natural language processing and traditional robot navigation by enabling researchers to fairly compare algorithms, revealing limitations of existing solutions and illuminating promising new directions. We believe the same approach can benefit social navigation. In this paper, we pave the road towards common, widely accessible, and repeatable benchmarking criteria to evaluate social robot navigation. Our contributions include (a) a definition of a socially navigating robot as one that respects the principles of safety, comfort, legibility, politeness, social competency, agent understanding, proactivity, and responsiveness to context, (b) guidelines for the use of metrics, development of scenarios, benchmarks, datasets, and simulators to evaluate social navigation, and (c) a design of a social navigation metrics framework to make it easier to compare results from different simulators, robots and datasets. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2306.16740v4-abstract-full').style.display = 'none'; document.getElementById('2306.16740v4-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, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 29 June, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">42 pages, 11 figures, 6 tables</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">ACM Class:</span> I.2.9 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2306.16526">arXiv:2306.16526</a> <span> [<a href="https://arxiv.org/pdf/2306.16526">pdf</a>, <a href="https://arxiv.org/format/2306.16526">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Information Retrieval">cs.IR</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Social and Information Networks">cs.SI</span> </div> </div> <p class="title is-5 mathjax"> Shilling Black-box Review-based Recommender Systems through Fake Review Generation </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Hung-Yun Chiang</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+Y">Yi-Syuan Chen</a>, <a href="/search/cs?searchtype=author&query=Song%2C+Y">Yun-Zhu Song</a>, <a href="/search/cs?searchtype=author&query=Shuai%2C+H">Hong-Han Shuai</a>, <a href="/search/cs?searchtype=author&query=Chang%2C+J+S">Jason S. Chang</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="2306.16526v1-abstract-short" style="display: inline;"> Review-Based Recommender Systems (RBRS) have attracted increasing research interest due to their ability to alleviate well-known cold-start problems. RBRS utilizes reviews to construct the user and items representations. However, in this paper, we argue that such a reliance on reviews may instead expose systems to the risk of being shilled. To explore this possibility, in this paper, we propose th… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2306.16526v1-abstract-full').style.display = 'inline'; document.getElementById('2306.16526v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2306.16526v1-abstract-full" style="display: none;"> Review-Based Recommender Systems (RBRS) have attracted increasing research interest due to their ability to alleviate well-known cold-start problems. RBRS utilizes reviews to construct the user and items representations. However, in this paper, we argue that such a reliance on reviews may instead expose systems to the risk of being shilled. To explore this possibility, in this paper, we propose the first generation-based model for shilling attacks against RBRSs. Specifically, we learn a fake review generator through reinforcement learning, which maliciously promotes items by forcing prediction shifts after adding generated reviews to the system. By introducing the auxiliary rewards to increase text fluency and diversity with the aid of pre-trained language models and aspect predictors, the generated reviews can be effective for shilling with high fidelity. Experimental results demonstrate that the proposed framework can successfully attack three different kinds of RBRSs on the Amazon corpus with three domains and Yelp corpus. Furthermore, human studies also show that the generated reviews are fluent and informative. Finally, equipped with Attack Review Generators (ARGs), RBRSs with adversarial training are much more robust to malicious reviews. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2306.16526v1-abstract-full').style.display = 'none'; document.getElementById('2306.16526v1-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 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/2306.08647">arXiv:2306.08647</a> <span> [<a href="https://arxiv.org/pdf/2306.08647">pdf</a>, <a href="https://arxiv.org/format/2306.08647">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"> Language to Rewards for Robotic Skill Synthesis </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Yu%2C+W">Wenhao Yu</a>, <a href="/search/cs?searchtype=author&query=Gileadi%2C+N">Nimrod Gileadi</a>, <a href="/search/cs?searchtype=author&query=Fu%2C+C">Chuyuan Fu</a>, <a href="/search/cs?searchtype=author&query=Kirmani%2C+S">Sean Kirmani</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+K">Kuang-Huei Lee</a>, <a href="/search/cs?searchtype=author&query=Arenas%2C+M+G">Montse Gonzalez Arenas</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+H+L">Hao-Tien Lewis Chiang</a>, <a href="/search/cs?searchtype=author&query=Erez%2C+T">Tom Erez</a>, <a href="/search/cs?searchtype=author&query=Hasenclever%2C+L">Leonard Hasenclever</a>, <a href="/search/cs?searchtype=author&query=Humplik%2C+J">Jan Humplik</a>, <a href="/search/cs?searchtype=author&query=Ichter%2C+B">Brian Ichter</a>, <a href="/search/cs?searchtype=author&query=Xiao%2C+T">Ted Xiao</a>, <a href="/search/cs?searchtype=author&query=Xu%2C+P">Peng Xu</a>, <a href="/search/cs?searchtype=author&query=Zeng%2C+A">Andy Zeng</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+T">Tingnan Zhang</a>, <a href="/search/cs?searchtype=author&query=Heess%2C+N">Nicolas Heess</a>, <a href="/search/cs?searchtype=author&query=Sadigh%2C+D">Dorsa Sadigh</a>, <a href="/search/cs?searchtype=author&query=Tan%2C+J">Jie Tan</a>, <a href="/search/cs?searchtype=author&query=Tassa%2C+Y">Yuval Tassa</a>, <a href="/search/cs?searchtype=author&query=Xia%2C+F">Fei Xia</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="2306.08647v2-abstract-short" style="display: inline;"> Large language models (LLMs) have demonstrated exciting progress in acquiring diverse new capabilities through in-context learning, ranging from logical reasoning to code-writing. Robotics researchers have also explored using LLMs to advance the capabilities of robotic control. However, since low-level robot actions are hardware-dependent and underrepresented in LLM training corpora, existing effo… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2306.08647v2-abstract-full').style.display = 'inline'; document.getElementById('2306.08647v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2306.08647v2-abstract-full" style="display: none;"> Large language models (LLMs) have demonstrated exciting progress in acquiring diverse new capabilities through in-context learning, ranging from logical reasoning to code-writing. Robotics researchers have also explored using LLMs to advance the capabilities of robotic control. However, since low-level robot actions are hardware-dependent and underrepresented in LLM training corpora, existing efforts in applying LLMs to robotics have largely treated LLMs as semantic planners or relied on human-engineered control primitives to interface with the robot. On the other hand, reward functions are shown to be flexible representations that can be optimized for control policies to achieve diverse tasks, while their semantic richness makes them suitable to be specified by LLMs. In this work, we introduce a new paradigm that harnesses this realization by utilizing LLMs to define reward parameters that can be optimized and accomplish variety of robotic tasks. Using reward as the intermediate interface generated by LLMs, we can effectively bridge the gap between high-level language instructions or corrections to low-level robot actions. Meanwhile, combining this with a real-time optimizer, MuJoCo MPC, empowers an interactive behavior creation experience where users can immediately observe the results and provide feedback to the system. To systematically evaluate the performance of our proposed method, we designed a total of 17 tasks for a simulated quadruped robot and a dexterous manipulator robot. We demonstrate that our proposed method reliably tackles 90% of the designed tasks, while a baseline using primitive skills as the interface with Code-as-policies achieves 50% of the tasks. We further validated our method on a real robot arm where complex manipulation skills such as non-prehensile pushing emerge through our interactive system. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2306.08647v2-abstract-full').style.display = 'none'; document.getElementById('2306.08647v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 16 June, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 14 June, 2023; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2023. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">https://language-to-reward.github.io/</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2212.03246">arXiv:2212.03246</a> <span> [<a href="https://arxiv.org/pdf/2212.03246">pdf</a>, <a href="https://arxiv.org/format/2212.03246">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> </div> </div> <p class="title is-5 mathjax"> MobileTL: On-device Transfer Learning with Inverted Residual Blocks </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Hung-Yueh Chiang</a>, <a href="/search/cs?searchtype=author&query=Frumkin%2C+N">Natalia Frumkin</a>, <a href="/search/cs?searchtype=author&query=Liang%2C+F">Feng Liang</a>, <a href="/search/cs?searchtype=author&query=Marculescu%2C+D">Diana Marculescu</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="2212.03246v2-abstract-short" style="display: inline;"> Transfer learning on edge is challenging due to on-device limited resources. Existing work addresses this issue by training a subset of parameters or adding model patches. Developed with inference in mind, Inverted Residual Blocks (IRBs) split a convolutional layer into depthwise and pointwise convolutions, leading to more stacking layers, e.g., convolution, normalization, and activation layers. T… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2212.03246v2-abstract-full').style.display = 'inline'; document.getElementById('2212.03246v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2212.03246v2-abstract-full" style="display: none;"> Transfer learning on edge is challenging due to on-device limited resources. Existing work addresses this issue by training a subset of parameters or adding model patches. Developed with inference in mind, Inverted Residual Blocks (IRBs) split a convolutional layer into depthwise and pointwise convolutions, leading to more stacking layers, e.g., convolution, normalization, and activation layers. Though they are efficient for inference, IRBs require that additional activation maps are stored in memory for training weights for convolution layers and scales for normalization layers. As a result, their high memory cost prohibits training IRBs on resource-limited edge devices, and making them unsuitable in the context of transfer learning. To address this issue, we present MobileTL, a memory and computationally efficient on-device transfer learning method for models built with IRBs. MobileTL trains the shifts for internal normalization layers to avoid storing activation maps for the backward pass. Also, MobileTL approximates the backward computation of the activation layer (e.g., Hard-Swish and ReLU6) as a signed function which enables storing a binary mask instead of activation maps for the backward pass. MobileTL fine-tunes a few top blocks (close to output) rather than propagating the gradient through the whole network to reduce the computation cost. Our method reduces memory usage by 46% and 53% for MobileNetV2 and V3 IRBs, respectively. For MobileNetV3, we observe a 36% reduction in floating-point operations (FLOPs) when fine-tuning 5 blocks, while only incurring a 0.6% accuracy reduction on CIFAR10. Extensive experiments on multiple datasets demonstrate that our method is Pareto-optimal (best accuracy under given hardware constraints) compared to prior work in transfer learning for edge devices. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2212.03246v2-abstract-full').style.display = 'none'; document.getElementById('2212.03246v2-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 April, 2023; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 5 December, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> December 2022. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2202.09686">arXiv:2202.09686</a> <span> [<a href="https://arxiv.org/pdf/2202.09686">pdf</a>, <a href="https://arxiv.org/format/2202.09686">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Computational Geometry">cs.CG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Graphics">cs.GR</span> </div> </div> <p class="title is-5 mathjax"> Local Decomposition of Hexahedral Singular Nodes into Singular Curves </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Zhang%2C+P">Paul Zhang</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+J+H">Judy Hsin-Hui Chiang</a>, <a href="/search/cs?searchtype=author&query=Xinyi"> Xinyi</a>, <a href="/search/cs?searchtype=author&query=Fan"> Fan</a>, <a href="/search/cs?searchtype=author&query=Mundilova%2C+K">Klara Mundilova</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="2202.09686v2-abstract-short" style="display: inline;"> Hexahedral (hex) meshing is a long studied topic in geometry processing with many fascinating and challenging associated problems. Hex meshes vary in complexity from structured to unstructured depending on application or domain of interest. Fully structured meshes require that all interior mesh edges are adjacent to exactly four hexes. Edges not satisfying this criteria are considered singular and… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2202.09686v2-abstract-full').style.display = 'inline'; document.getElementById('2202.09686v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2202.09686v2-abstract-full" style="display: none;"> Hexahedral (hex) meshing is a long studied topic in geometry processing with many fascinating and challenging associated problems. Hex meshes vary in complexity from structured to unstructured depending on application or domain of interest. Fully structured meshes require that all interior mesh edges are adjacent to exactly four hexes. Edges not satisfying this criteria are considered singular and indicate an unstructured hex mesh. Singular edges join together into singular curves that either form closed cycles, end on the mesh boundary, or end at a singular node, a complex junction of more than two singular curves. While all hex meshes with singularities are unstructured, those with more complex singular nodes tend to have more distorted elements and smaller scaled Jacobian values. In this work, we study the topology of singular nodes. We show that all eight of the most common singular nodes are decomposable into just singular curves. We further show that all singular nodes, regardless of edge valence, are locally decomposable. Finally we demonstrate these decompositions on hex meshes, thereby decreasing their distortion and converting all singular nodes into singular curves. With this decomposition, the enigmatic complexity of 3D singular nodes becomes effectively 2D. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2202.09686v2-abstract-full').style.display = 'none'; document.getElementById('2202.09686v2-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 September, 2024; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 19 February, 2022; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2022. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2111.05691">arXiv:2111.05691</a> <span> [<a href="https://arxiv.org/pdf/2111.05691">pdf</a>, <a href="https://arxiv.org/format/2111.05691">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Audio and Speech Processing">eess.AS</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Artificial Intelligence">cs.AI</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Sound">cs.SD</span> </div> </div> <p class="title is-5 mathjax"> HASA-net: A non-intrusive hearing-aid speech assessment network </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Hsin-Tien Chiang</a>, <a href="/search/cs?searchtype=author&query=Wu%2C+Y">Yi-Chiao Wu</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+C">Cheng Yu</a>, <a href="/search/cs?searchtype=author&query=Toda%2C+T">Tomoki Toda</a>, <a href="/search/cs?searchtype=author&query=Wang%2C+H">Hsin-Min Wang</a>, <a href="/search/cs?searchtype=author&query=Hu%2C+Y">Yih-Chun Hu</a>, <a href="/search/cs?searchtype=author&query=Tsao%2C+Y">Yu Tsao</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2111.05691v1-abstract-short" style="display: inline;"> Without the need of a clean reference, non-intrusive speech assessment methods have caught great attention for objective evaluations. Recently, deep neural network (DNN) models have been applied to build non-intrusive speech assessment approaches and confirmed to provide promising performance. However, most DNN-based approaches are designed for normal-hearing listeners without considering hearing-… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2111.05691v1-abstract-full').style.display = 'inline'; document.getElementById('2111.05691v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2111.05691v1-abstract-full" style="display: none;"> Without the need of a clean reference, non-intrusive speech assessment methods have caught great attention for objective evaluations. Recently, deep neural network (DNN) models have been applied to build non-intrusive speech assessment approaches and confirmed to provide promising performance. However, most DNN-based approaches are designed for normal-hearing listeners without considering hearing-loss factors. In this study, we propose a DNN-based hearing aid speech assessment network (HASA-Net), formed by a bidirectional long short-term memory (BLSTM) model, to predict speech quality and intelligibility scores simultaneously according to input speech signals and specified hearing-loss patterns. To the best of our knowledge, HASA-Net is the first work to incorporate quality and intelligibility assessments utilizing a unified DNN-based non-intrusive model for hearing aids. Experimental results show that the predicted speech quality and intelligibility scores of HASA-Net are highly correlated to two well-known intrusive hearing-aid evaluation metrics, hearing aid speech quality index (HASQI) and hearing aid speech perception index (HASPI), respectively. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2111.05691v1-abstract-full').style.display = 'none'; document.getElementById('2111.05691v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 10 November, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 2021. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2108.08189">arXiv:2108.08189</a> <span> [<a href="https://arxiv.org/pdf/2108.08189">pdf</a>, <a href="https://arxiv.org/format/2108.08189">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computer Vision and Pattern Recognition">cs.CV</span> </div> </div> <p class="title is-5 mathjax"> FOX-NAS: Fast, On-device and Explainable Neural Architecture Search </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Liu%2C+C">Chia-Hsiang Liu</a>, <a href="/search/cs?searchtype=author&query=Han%2C+Y">Yu-Shin Han</a>, <a href="/search/cs?searchtype=author&query=Sung%2C+Y">Yuan-Yao Sung</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+Y">Yi Lee</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Hung-Yueh Chiang</a>, <a href="/search/cs?searchtype=author&query=Wu%2C+K">Kai-Chiang Wu</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2108.08189v1-abstract-short" style="display: inline;"> Neural architecture search can discover neural networks with good performance, and One-Shot approaches are prevalent. One-Shot approaches typically require a supernet with weight sharing and predictors that predict the performance of architecture. However, the previous methods take much time to generate performance predictors thus are inefficient. To this end, we propose FOX-NAS that consists of f… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2108.08189v1-abstract-full').style.display = 'inline'; document.getElementById('2108.08189v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2108.08189v1-abstract-full" style="display: none;"> Neural architecture search can discover neural networks with good performance, and One-Shot approaches are prevalent. One-Shot approaches typically require a supernet with weight sharing and predictors that predict the performance of architecture. However, the previous methods take much time to generate performance predictors thus are inefficient. To this end, we propose FOX-NAS that consists of fast and explainable predictors based on simulated annealing and multivariate regression. Our method is quantization-friendly and can be efficiently deployed to the edge. The experiments on different hardware show that FOX-NAS models outperform some other popular neural network architectures. For example, FOX-NAS matches MobileNetV2 and EfficientNet-Lite0 accuracy with 240% and 40% less latency on the edge CPU. FOX-NAS is the 3rd place winner of the 2020 Low-Power Computer Vision Challenge (LPCVC), DSP classification track. See all evaluation results at https://lpcv.ai/competitions/2020. Search code and pre-trained models are released at https://github.com/great8nctu/FOX-NAS. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2108.08189v1-abstract-full').style.display = 'none'; document.getElementById('2108.08189v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 14 August, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> August 2021. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Accepted by ICCV 2021 Low-Power Computer Vision Workshop</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2106.11548">arXiv:2106.11548</a> <span> [<a href="https://arxiv.org/pdf/2106.11548">pdf</a>, <a href="https://arxiv.org/format/2106.11548">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"> Adaptive Learning Rate and Momentum for Training Deep Neural Networks </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Hao%2C+Z">Zhiyong Hao</a>, <a href="/search/cs?searchtype=author&query=Jiang%2C+Y">Yixuan Jiang</a>, <a href="/search/cs?searchtype=author&query=Yu%2C+H">Huihua Yu</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Hsiao-Dong Chiang</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2106.11548v2-abstract-short" style="display: inline;"> Recent progress on deep learning relies heavily on the quality and efficiency of training algorithms. In this paper, we develop a fast training method motivated by the nonlinear Conjugate Gradient (CG) framework. We propose the Conjugate Gradient with Quadratic line-search (CGQ) method. On the one hand, a quadratic line-search determines the step size according to current loss landscape. On the ot… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2106.11548v2-abstract-full').style.display = 'inline'; document.getElementById('2106.11548v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2106.11548v2-abstract-full" style="display: none;"> Recent progress on deep learning relies heavily on the quality and efficiency of training algorithms. In this paper, we develop a fast training method motivated by the nonlinear Conjugate Gradient (CG) framework. We propose the Conjugate Gradient with Quadratic line-search (CGQ) method. On the one hand, a quadratic line-search determines the step size according to current loss landscape. On the other hand, the momentum factor is dynamically updated in computing the conjugate gradient parameter (like Polak-Ribiere). Theoretical results to ensure the convergence of our method in strong convex settings is developed. And experiments in image classification datasets show that our method yields faster convergence than other local solvers and has better generalization capability (test set accuracy). One major advantage of the paper method is that tedious hand tuning of hyperparameters like the learning rate and momentum is avoided. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2106.11548v2-abstract-full').style.display = 'none'; document.getElementById('2106.11548v2-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 July, 2021; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 22 June, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2021. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">accepted to ECML PKDD 2021</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2106.08417">arXiv:2106.08417</a> <span> [<a href="https://arxiv.org/pdf/2106.08417">pdf</a>, <a href="https://arxiv.org/format/2106.08417">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="Robotics">cs.RO</span> </div> </div> <p class="title is-5 mathjax"> Scene Transformer: A unified architecture for predicting multiple agent trajectories </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Ngiam%2C+J">Jiquan Ngiam</a>, <a href="/search/cs?searchtype=author&query=Caine%2C+B">Benjamin Caine</a>, <a href="/search/cs?searchtype=author&query=Vasudevan%2C+V">Vijay Vasudevan</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+Z">Zhengdong Zhang</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+H+L">Hao-Tien Lewis Chiang</a>, <a href="/search/cs?searchtype=author&query=Ling%2C+J">Jeffrey Ling</a>, <a href="/search/cs?searchtype=author&query=Roelofs%2C+R">Rebecca Roelofs</a>, <a href="/search/cs?searchtype=author&query=Bewley%2C+A">Alex Bewley</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+C">Chenxi Liu</a>, <a href="/search/cs?searchtype=author&query=Venugopal%2C+A">Ashish Venugopal</a>, <a href="/search/cs?searchtype=author&query=Weiss%2C+D">David Weiss</a>, <a href="/search/cs?searchtype=author&query=Sapp%2C+B">Ben Sapp</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+Z">Zhifeng Chen</a>, <a href="/search/cs?searchtype=author&query=Shlens%2C+J">Jonathon Shlens</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2106.08417v3-abstract-short" style="display: inline;"> Predicting the motion of multiple agents is necessary for planning in dynamic environments. This task is challenging for autonomous driving since agents (e.g. vehicles and pedestrians) and their associated behaviors may be diverse and influence one another. Most prior work have focused on predicting independent futures for each agent based on all past motion, and planning against these independent… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2106.08417v3-abstract-full').style.display = 'inline'; document.getElementById('2106.08417v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2106.08417v3-abstract-full" style="display: none;"> Predicting the motion of multiple agents is necessary for planning in dynamic environments. This task is challenging for autonomous driving since agents (e.g. vehicles and pedestrians) and their associated behaviors may be diverse and influence one another. Most prior work have focused on predicting independent futures for each agent based on all past motion, and planning against these independent predictions. However, planning against independent predictions can make it challenging to represent the future interaction possibilities between different agents, leading to sub-optimal planning. In this work, we formulate a model for predicting the behavior of all agents jointly, producing consistent futures that account for interactions between agents. Inspired by recent language modeling approaches, we use a masking strategy as the query to our model, enabling one to invoke a single model to predict agent behavior in many ways, such as potentially conditioned on the goal or full future trajectory of the autonomous vehicle or the behavior of other agents in the environment. Our model architecture employs attention to combine features across road elements, agent interactions, and time steps. We evaluate our approach on autonomous driving datasets for both marginal and joint motion prediction, and achieve state of the art performance across two popular datasets. Through combining a scene-centric approach, agent permutation equivariant model, and a sequence masking strategy, we show that our model can unify a variety of motion prediction tasks from joint motion predictions to conditioned prediction. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2106.08417v3-abstract-full').style.display = 'none'; document.getElementById('2106.08417v3-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 March, 2022; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 15 June, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2021. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">ICLR 2022</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2106.01870">arXiv:2106.01870</a> <span> [<a href="https://arxiv.org/pdf/2106.01870">pdf</a>, <a href="https://arxiv.org/format/2106.01870">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="Computational Engineering, Finance, and Science">cs.CE</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Formal Languages and Automata Theory">cs.FL</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computer Science and Game Theory">cs.GT</span> </div> </div> <p class="title is-5 mathjax"> Maximizing Extractable Value from Automated Market Makers </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Bartoletti%2C+M">Massimo Bartoletti</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+J+H">James Hsin-yu Chiang</a>, <a href="/search/cs?searchtype=author&query=Lluch-Lafuente%2C+A">Alberto Lluch-Lafuente</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="2106.01870v4-abstract-short" style="display: inline;"> Automated Market Makers (AMMs) are decentralized applications that allow users to exchange crypto-tokens without the need for a matching exchange order. AMMs are one of the most successful DeFi use cases: indeed, major AMM platforms process a daily volume of transactions worth USD billions. Despite their popularity, AMMs are well-known to suffer from transaction-ordering issues: adversaries can in… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2106.01870v4-abstract-full').style.display = 'inline'; document.getElementById('2106.01870v4-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2106.01870v4-abstract-full" style="display: none;"> Automated Market Makers (AMMs) are decentralized applications that allow users to exchange crypto-tokens without the need for a matching exchange order. AMMs are one of the most successful DeFi use cases: indeed, major AMM platforms process a daily volume of transactions worth USD billions. Despite their popularity, AMMs are well-known to suffer from transaction-ordering issues: adversaries can influence the ordering of user transactions, and possibly front-run them with their own, to extract value from AMMs, to the detriment of users. We devise an effective procedure to construct a strategy through which an adversary can maximize the value extracted from user transactions. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2106.01870v4-abstract-full').style.display = 'none'; document.getElementById('2106.01870v4-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 July, 2022; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 2 June, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 2021. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">FC'22 proceedings version with proofs in appendix</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">MSC Class:</span> 68N30 <span class="has-text-black-bis has-text-weight-semibold">ACM Class:</span> I.6.4 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2104.04687">arXiv:2104.04687</a> <span> [<a href="https://arxiv.org/pdf/2104.04687">pdf</a>, <a href="https://arxiv.org/format/2104.04687">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"> Learning from 2D: Contrastive Pixel-to-Point Knowledge Transfer for 3D Pretraining </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Liu%2C+Y">Yueh-Cheng Liu</a>, <a href="/search/cs?searchtype=author&query=Huang%2C+Y">Yu-Kai Huang</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Hung-Yueh Chiang</a>, <a href="/search/cs?searchtype=author&query=Su%2C+H">Hung-Ting Su</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+Z">Zhe-Yu Liu</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+C">Chin-Tang Chen</a>, <a href="/search/cs?searchtype=author&query=Tseng%2C+C">Ching-Yu Tseng</a>, <a href="/search/cs?searchtype=author&query=Hsu%2C+W+H">Winston H. Hsu</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="2104.04687v3-abstract-short" style="display: inline;"> Most 3D neural networks are trained from scratch owing to the lack of large-scale labeled 3D datasets. In this paper, we present a novel 3D pretraining method by leveraging 2D networks learned from rich 2D datasets. We propose the contrastive pixel-to-point knowledge transfer to effectively utilize the 2D information by mapping the pixel-level and point-level features into the same embedding space… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2104.04687v3-abstract-full').style.display = 'inline'; document.getElementById('2104.04687v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2104.04687v3-abstract-full" style="display: none;"> Most 3D neural networks are trained from scratch owing to the lack of large-scale labeled 3D datasets. In this paper, we present a novel 3D pretraining method by leveraging 2D networks learned from rich 2D datasets. We propose the contrastive pixel-to-point knowledge transfer to effectively utilize the 2D information by mapping the pixel-level and point-level features into the same embedding space. Due to the heterogeneous nature between 2D and 3D networks, we introduce the back-projection function to align the features between 2D and 3D to make the transfer possible. Additionally, we devise an upsampling feature projection layer to increase the spatial resolution of high-level 2D feature maps, which enables learning fine-grained 3D representations. With a pretrained 2D network, the proposed pretraining process requires no additional 2D or 3D labeled data, further alleviating the expensive 3D data annotation cost. To the best of our knowledge, we are the first to exploit existing 2D trained weights to pretrain 3D deep neural networks. Our intensive experiments show that the 3D models pretrained with 2D knowledge boost the performances of 3D networks across various real-world 3D downstream tasks. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2104.04687v3-abstract-full').style.display = 'none'; document.getElementById('2104.04687v3-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 December, 2021; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 10 April, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> April 2021. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2102.11350">arXiv:2102.11350</a> <span> [<a href="https://arxiv.org/pdf/2102.11350">pdf</a>, <a href="https://arxiv.org/format/2102.11350">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Formal Languages and Automata Theory">cs.FL</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="Computer Science and Game Theory">cs.GT</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/10.46298/lmcs-18(4:12)2022">10.46298/lmcs-18(4:12)2022 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> A theory of Automated Market Makers in DeFi </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Bartoletti%2C+M">Massimo Bartoletti</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+J+H">James Hsin-yu Chiang</a>, <a href="/search/cs?searchtype=author&query=Lluch-Lafuente%2C+A">Alberto Lluch-Lafuente</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="2102.11350v7-abstract-short" style="display: inline;"> Automated market makers (AMMs) are one of the most prominent decentralized finance (DeFi) applications. AMMs allow users to trade different types of crypto-tokens, without the need to find a counter-party. There are several implementations and models for AMMs, featuring a variety of sophisticated economic mechanisms. We present a theory of AMMs. The core of our theory is an abstract operational mo… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2102.11350v7-abstract-full').style.display = 'inline'; document.getElementById('2102.11350v7-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2102.11350v7-abstract-full" style="display: none;"> Automated market makers (AMMs) are one of the most prominent decentralized finance (DeFi) applications. AMMs allow users to trade different types of crypto-tokens, without the need to find a counter-party. There are several implementations and models for AMMs, featuring a variety of sophisticated economic mechanisms. We present a theory of AMMs. The core of our theory is an abstract operational model of the interactions between users and AMMs, which can be concretised by instantiating the economic mechanisms. We exploit our theory to formally prove a set of fundamental properties of AMMs, characterizing both structural and economic aspects. We do this by abstracting from the actual economic mechanisms used in implementations, and identifying sufficient conditions which ensure the relevant properties. Notably, we devise a general solution to the arbitrage problem, the main game-theoretic foundation behind the economic mechanisms of AMMs. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2102.11350v7-abstract-full').style.display = 'none'; document.getElementById('2102.11350v7-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 December, 2022; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 22 February, 2021; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2021. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">MSC Class:</span> 68N30 <span class="has-text-black-bis has-text-weight-semibold">ACM Class:</span> I.6.4 </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> Logical Methods in Computer Science, Volume 18, Issue 4 (December 19, 2022) lmcs:8955 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2012.13230">arXiv:2012.13230</a> <span> [<a href="https://arxiv.org/pdf/2012.13230">pdf</a>, <a href="https://arxiv.org/ps/2012.13230">ps</a>, <a href="https://arxiv.org/format/2012.13230">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="Computer Science and Game Theory">cs.GT</span> <span class="tag is-small is-grey tooltip is-tooltip-top" data-tooltip="General Finance">q-fin.GN</span> </div> </div> <p class="title is-5 mathjax"> SoK: Lending Pools in Decentralized Finance </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Bartoletti%2C+M">Massimo Bartoletti</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+J+H">James Hsin-yu Chiang</a>, <a href="/search/cs?searchtype=author&query=Lluch-Lafuente%2C+A">Alberto Lluch-Lafuente</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="2012.13230v1-abstract-short" style="display: inline;"> Lending pools are decentralized applications which allow mutually untrusted users to lend and borrow crypto-assets. These applications feature complex, highly parametric incentive mechanisms to equilibrate the loan market. This complexity makes the behaviour of lending pools difficult to understand and to predict: indeed, ineffective incentives and attacks could potentially lead to emergent unwant… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2012.13230v1-abstract-full').style.display = 'inline'; document.getElementById('2012.13230v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2012.13230v1-abstract-full" style="display: none;"> Lending pools are decentralized applications which allow mutually untrusted users to lend and borrow crypto-assets. These applications feature complex, highly parametric incentive mechanisms to equilibrate the loan market. This complexity makes the behaviour of lending pools difficult to understand and to predict: indeed, ineffective incentives and attacks could potentially lead to emergent unwanted behaviours. Reasoning about lending pools is made even harder by the lack of executable models of their behaviour: to precisely understand how users interact with lending pools, eventually one has to inspect their implementations, where the incentive mechanisms are intertwined with low-level implementation details. Further, the variety of existing implementations makes it difficult to distill the common aspects of lending pools. We systematize the existing knowledge about lending pools, leveraging a new formal model of interactions with users, which reflects the archetypal features of mainstream implementations. This enables us to prove some general properties of lending pools, such as the correct handling of funds, and to precisely describe vulnerabilities and attacks. We also discuss the role of lending pools in the broader context of decentralized finance. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2012.13230v1-abstract-full').style.display = 'none'; document.getElementById('2012.13230v1-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 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">20 pages. Under submission</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">MSC Class:</span> 68N30 <span class="has-text-black-bis has-text-weight-semibold">ACM Class:</span> I.6.4 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2006.04646">arXiv:2006.04646</a> <span> [<a href="https://arxiv.org/pdf/2006.04646">pdf</a>, <a href="https://arxiv.org/format/2006.04646">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Social and Information Networks">cs.SI</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="Physics and Society">physics.soc-ph</span> </div> </div> <p class="title is-5 mathjax"> Continuous Learning and Inference of Individual Probability of SARS-CoV-2 Infection Based on Interaction Data </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Liu%2C+S">Shangching Liu</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+K">Koyun Liu</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Hwaihai Chiang</a>, <a href="/search/cs?searchtype=author&query=Zhang%2C+J">Jianwei Zhang</a>, <a href="/search/cs?searchtype=author&query=Chang%2C+T">Tsungyao Chang</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="2006.04646v3-abstract-short" style="display: inline;"> This study presents a new approach to determine the likelihood of asymptomatic carriers of the SARS-CoV-2 virus by using interaction-based continuous learning and inference of individual probability (CLIIP) for contagious ranking. This approach is developed based on an individual directed graph (IDG), using multi-layer bidirectional path tracking and inference searching. The IDG is determined by t… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2006.04646v3-abstract-full').style.display = 'inline'; document.getElementById('2006.04646v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2006.04646v3-abstract-full" style="display: none;"> This study presents a new approach to determine the likelihood of asymptomatic carriers of the SARS-CoV-2 virus by using interaction-based continuous learning and inference of individual probability (CLIIP) for contagious ranking. This approach is developed based on an individual directed graph (IDG), using multi-layer bidirectional path tracking and inference searching. The IDG is determined by the appearance timeline and spatial data that can adapt over time. Additionally, the approach takes into consideration the incubation period and several features that can represent real-world circumstances, such as the number of asymptomatic carriers present. After each update of confirmed cases, the model collects the interaction features and infers the individual person's probability of getting infected using the status of the surrounding people. The CLIIP approach is validated using the individualized bidirectional SEIR model to simulate the contagion process. Compared to traditional contact tracing methods, our approach significantly reduces the screening and quarantine required to search for the potential asymptomatic virus carriers by as much as 94%. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2006.04646v3-abstract-full').style.display = 'none'; document.getElementById('2006.04646v3-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 January, 2021; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 8 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">15 pages,6 figures,2 tables</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">MSC Class:</span> 68T07(Primary); 68T09(Secondary) </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/2001.11190">arXiv:2001.11190</a> <span> [<a href="https://arxiv.org/pdf/2001.11190">pdf</a>, <a href="https://arxiv.org/format/2001.11190">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="Robotics">cs.RO</span> </div> </div> <p class="title is-5 mathjax"> 2018 Robotic Scene Segmentation Challenge </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Allan%2C+M">Max Allan</a>, <a href="/search/cs?searchtype=author&query=Kondo%2C+S">Satoshi Kondo</a>, <a href="/search/cs?searchtype=author&query=Bodenstedt%2C+S">Sebastian Bodenstedt</a>, <a href="/search/cs?searchtype=author&query=Leger%2C+S">Stefan Leger</a>, <a href="/search/cs?searchtype=author&query=Kadkhodamohammadi%2C+R">Rahim Kadkhodamohammadi</a>, <a href="/search/cs?searchtype=author&query=Luengo%2C+I">Imanol Luengo</a>, <a href="/search/cs?searchtype=author&query=Fuentes%2C+F">Felix Fuentes</a>, <a href="/search/cs?searchtype=author&query=Flouty%2C+E">Evangello Flouty</a>, <a href="/search/cs?searchtype=author&query=Mohammed%2C+A">Ahmed Mohammed</a>, <a href="/search/cs?searchtype=author&query=Pedersen%2C+M">Marius Pedersen</a>, <a href="/search/cs?searchtype=author&query=Kori%2C+A">Avinash Kori</a>, <a href="/search/cs?searchtype=author&query=Alex%2C+V">Varghese Alex</a>, <a href="/search/cs?searchtype=author&query=Krishnamurthi%2C+G">Ganapathy Krishnamurthi</a>, <a href="/search/cs?searchtype=author&query=Rauber%2C+D">David Rauber</a>, <a href="/search/cs?searchtype=author&query=Mendel%2C+R">Robert Mendel</a>, <a href="/search/cs?searchtype=author&query=Palm%2C+C">Christoph Palm</a>, <a href="/search/cs?searchtype=author&query=Bano%2C+S">Sophia Bano</a>, <a href="/search/cs?searchtype=author&query=Saibro%2C+G">Guinther Saibro</a>, <a href="/search/cs?searchtype=author&query=Shih%2C+C">Chi-Sheng Shih</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Hsun-An Chiang</a>, <a href="/search/cs?searchtype=author&query=Zhuang%2C+J">Juntang Zhuang</a>, <a href="/search/cs?searchtype=author&query=Yang%2C+J">Junlin Yang</a>, <a href="/search/cs?searchtype=author&query=Iglovikov%2C+V">Vladimir Iglovikov</a>, <a href="/search/cs?searchtype=author&query=Dobrenkii%2C+A">Anton Dobrenkii</a>, <a href="/search/cs?searchtype=author&query=Reddiboina%2C+M">Madhu Reddiboina</a> , et al. (16 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="2001.11190v3-abstract-short" style="display: inline;"> In 2015 we began a sub-challenge at the EndoVis workshop at MICCAI in Munich using endoscope images of ex-vivo tissue with automatically generated annotations from robot forward kinematics and instrument CAD models. However, the limited background variation and simple motion rendered the dataset uninformative in learning about which techniques would be suitable for segmentation in real surgery. In… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2001.11190v3-abstract-full').style.display = 'inline'; document.getElementById('2001.11190v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="2001.11190v3-abstract-full" style="display: none;"> In 2015 we began a sub-challenge at the EndoVis workshop at MICCAI in Munich using endoscope images of ex-vivo tissue with automatically generated annotations from robot forward kinematics and instrument CAD models. However, the limited background variation and simple motion rendered the dataset uninformative in learning about which techniques would be suitable for segmentation in real surgery. In 2017, at the same workshop in Quebec we introduced the robotic instrument segmentation dataset with 10 teams participating in the challenge to perform binary, articulating parts and type segmentation of da Vinci instruments. This challenge included realistic instrument motion and more complex porcine tissue as background and was widely addressed with modifications on U-Nets and other popular CNN architectures. In 2018 we added to the complexity by introducing a set of anatomical objects and medical devices to the segmented classes. To avoid over-complicating the challenge, we continued with porcine data which is dramatically simpler than human tissue due to the lack of fatty tissue occluding many organs. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('2001.11190v3-abstract-full').style.display = 'none'; document.getElementById('2001.11190v3-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 2 August, 2020; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 30 January, 2020; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> January 2020. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1910.13538">arXiv:1910.13538</a> <span> [<a href="https://arxiv.org/pdf/1910.13538">pdf</a>, <a href="https://arxiv.org/ps/1910.13538">ps</a>, <a href="https://arxiv.org/format/1910.13538">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="Signal Processing">eess.SP</span> </div> </div> <p class="title is-5 mathjax"> Machine-Learning Beam Tracking and Weight Optimization for mmWave Multi-UAV Links </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Hsiao-Lan Chiang</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+K">Kwang-Cheng Chen</a>, <a href="/search/cs?searchtype=author&query=Rave%2C+W">Wolfgang Rave</a>, <a href="/search/cs?searchtype=author&query=Marandi%2C+M+K">Mostafa Khalili Marandi</a>, <a href="/search/cs?searchtype=author&query=Fettweis%2C+G">Gerhard Fettweis</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="1910.13538v1-abstract-short" style="display: inline;"> Millimeter-wave (mmWave) hybrid analog-digital beamforming is a promising approach to satisfy the low-latency constraint in multiple unmanned aerial vehicles (UAVs) systems, which serve as network infrastructure for flexible deployment. However, in highly dynamic multi-UAV environments, analog beam tracking becomes a critical challenge. The overhead of additional pilot transmission at the price of… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1910.13538v1-abstract-full').style.display = 'inline'; document.getElementById('1910.13538v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1910.13538v1-abstract-full" style="display: none;"> Millimeter-wave (mmWave) hybrid analog-digital beamforming is a promising approach to satisfy the low-latency constraint in multiple unmanned aerial vehicles (UAVs) systems, which serve as network infrastructure for flexible deployment. However, in highly dynamic multi-UAV environments, analog beam tracking becomes a critical challenge. The overhead of additional pilot transmission at the price of spectral efficiency is shown necessary to achieve high resilience in operation. An efficient method to deal with high dynamics of UAVs applies machine learning, particularly Q-learning, to analog beam tracking. The proposed Q-learning-based beam tracking scheme uses current/past observations to design rewards from environments to facilitate prediction, which significantly increases the efficiency of data transmission and beam switching. Given the selected analog beams, the goal of digital beamforming is to maximize the SINR. The received pilot signals are utilized to approximate the desired signal and interference power, which yield the SINR measurements as well as the optimal digital weights. Since the selected analog beams based on the received power do not guarantee the hybrid beamforming achieving the maximization SINR, we therefore reserve additional analog beams as candidates during the beam tracking. The combination of analog beams with their digital weights achieving the maximum SINR consequently provides the optimal solution to the hybrid beamforming. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1910.13538v1-abstract-full').style.display = 'none'; document.getElementById('1910.13538v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 29 October, 2019; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> October 2019. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1909.10153">arXiv:1909.10153</a> <span> [<a href="https://arxiv.org/pdf/1909.10153">pdf</a>, <a href="https://arxiv.org/format/1909.10153">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="Image and Video Processing">eess.IV</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/10.1117/12.2081310">10.1117/12.2081310 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Smooth Extrapolation of Unknown Anatomy via Statistical Shape Models </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Grupp%2C+R">Robert Grupp</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Hsin-Hong Chiang</a>, <a href="/search/cs?searchtype=author&query=Otake%2C+Y">Yoshito Otake</a>, <a href="/search/cs?searchtype=author&query=Murphy%2C+R">Ryan Murphy</a>, <a href="/search/cs?searchtype=author&query=Gordon%2C+C">Chad Gordon</a>, <a href="/search/cs?searchtype=author&query=Armand%2C+M">Mehran Armand</a>, <a href="/search/cs?searchtype=author&query=Taylor%2C+R">Russell Taylor</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="1909.10153v1-abstract-short" style="display: inline;"> Several methods to perform extrapolation of unknown anatomy were evaluated. The primary application is to enhance surgical procedures that may use partial medical images or medical images of incomplete anatomy. Le Fort-based, face-jaw-teeth transplant is one such procedure. From CT data of 36 skulls and 21 mandibles separate Statistical Shape Models of the anatomical surfaces were created. Using t… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1909.10153v1-abstract-full').style.display = 'inline'; document.getElementById('1909.10153v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1909.10153v1-abstract-full" style="display: none;"> Several methods to perform extrapolation of unknown anatomy were evaluated. The primary application is to enhance surgical procedures that may use partial medical images or medical images of incomplete anatomy. Le Fort-based, face-jaw-teeth transplant is one such procedure. From CT data of 36 skulls and 21 mandibles separate Statistical Shape Models of the anatomical surfaces were created. Using the Statistical Shape Models, incomplete surfaces were projected to obtain complete surface estimates. The surface estimates exhibit non-zero error in regions where the true surface is known; it is desirable to keep the true surface and seamlessly merge the estimated unknown surface. Existing extrapolation techniques produce non-smooth transitions from the true surface to the estimated surface, resulting in additional error and a less aesthetically pleasing result. The three extrapolation techniques evaluated were: copying and pasting of the surface estimate (non-smooth baseline), a feathering between the patient surface and surface estimate, and an estimate generated via a Thin Plate Spline trained from displacements between the surface estimate and corresponding vertices of the known patient surface. Feathering and Thin Plate Spline approaches both yielded smooth transitions. However, feathering corrupted known vertex values. Leave-one-out analyses were conducted, with 5% to 50% of known anatomy removed from the left-out patient and estimated via the proposed approaches. The Thin Plate Spline approach yielded smaller errors than the other two approaches, with an average vertex error improvement of 1.46 mm and 1.38 mm for the skull and mandible respectively, over the baseline approach. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1909.10153v1-abstract-full').style.display = 'none'; document.getElementById('1909.10153v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 23 September, 2019; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 2019. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">SPIE Medical Imaging Conference 2015 Paper</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> In Medical Imaging 2015: Image-Guided Procedures, Robotic Interventions, and Modeling 2015 Mar 18 (Vol. 9415, p. 941524). International Society for Optics and Photonics </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1908.00478">arXiv:1908.00478</a> <span> [<a href="https://arxiv.org/pdf/1908.00478">pdf</a>, <a href="https://arxiv.org/format/1908.00478">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"> A Unified Point-Based Framework for 3D Segmentation </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Hung-Yueh Chiang</a>, <a href="/search/cs?searchtype=author&query=Lin%2C+Y">Yen-Liang Lin</a>, <a href="/search/cs?searchtype=author&query=Liu%2C+Y">Yueh-Cheng Liu</a>, <a href="/search/cs?searchtype=author&query=Hsu%2C+W+H">Winston H. Hsu</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="1908.00478v4-abstract-short" style="display: inline;"> 3D point cloud segmentation remains challenging for structureless and textureless regions. We present a new unified point-based framework for 3D point cloud segmentation that effectively optimizes pixel-level features, geometrical structures and global context priors of an entire scene. By back-projecting 2D image features into 3D coordinates, our network learns 2D textural appearance and 3D struc… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1908.00478v4-abstract-full').style.display = 'inline'; document.getElementById('1908.00478v4-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1908.00478v4-abstract-full" style="display: none;"> 3D point cloud segmentation remains challenging for structureless and textureless regions. We present a new unified point-based framework for 3D point cloud segmentation that effectively optimizes pixel-level features, geometrical structures and global context priors of an entire scene. By back-projecting 2D image features into 3D coordinates, our network learns 2D textural appearance and 3D structural features in a unified framework. In addition, we investigate a global context prior to obtain a better prediction. We evaluate our framework on ScanNet online benchmark and show that our method outperforms several state-of-the-art approaches. We explore synthesizing camera poses in 3D reconstructed scenes for achieving higher performance. In-depth analysis on feature combinations and synthetic camera pose verify that features from different modalities benefit each other and dense camera pose sampling further improves the segmentation results. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1908.00478v4-abstract-full').style.display = 'none'; document.getElementById('1908.00478v4-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 18 August, 2019; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 1 August, 2019; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> August 2019. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1907.04799">arXiv:1907.04799</a> <span> [<a href="https://arxiv.org/pdf/1907.04799">pdf</a>, <a href="https://arxiv.org/format/1907.04799">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"> RL-RRT: Kinodynamic Motion Planning via Learning Reachability Estimators from RL Policies </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chiang%2C+H+L">Hao-Tien Lewis Chiang</a>, <a href="/search/cs?searchtype=author&query=Hsu%2C+J">Jasmine Hsu</a>, <a href="/search/cs?searchtype=author&query=Fiser%2C+M">Marek Fiser</a>, <a href="/search/cs?searchtype=author&query=Tapia%2C+L">Lydia Tapia</a>, <a href="/search/cs?searchtype=author&query=Faust%2C+A">Aleksandra Faust</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="1907.04799v2-abstract-short" style="display: inline;"> This paper addresses two challenges facing sampling-based kinodynamic motion planning: a way to identify good candidate states for local transitions and the subsequent computationally intractable steering between these candidate states. Through the combination of sampling-based planning, a Rapidly Exploring Randomized Tree (RRT) and an efficient kinodynamic motion planner through machine learning,… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1907.04799v2-abstract-full').style.display = 'inline'; document.getElementById('1907.04799v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1907.04799v2-abstract-full" style="display: none;"> This paper addresses two challenges facing sampling-based kinodynamic motion planning: a way to identify good candidate states for local transitions and the subsequent computationally intractable steering between these candidate states. Through the combination of sampling-based planning, a Rapidly Exploring Randomized Tree (RRT) and an efficient kinodynamic motion planner through machine learning, we propose an efficient solution to long-range planning for kinodynamic motion planning. First, we use deep reinforcement learning to learn an obstacle-avoiding policy that maps a robot's sensor observations to actions, which is used as a local planner during planning and as a controller during execution. Second, we train a reachability estimator in a supervised manner, which predicts the RL policy's time to reach a state in the presence of obstacles. Lastly, we introduce RL-RRT that uses the RL policy as a local planner, and the reachability estimator as the distance function to bias tree-growth towards promising regions. We evaluate our method on three kinodynamic systems, including physical robot experiments. Results across all three robots tested indicate that RL-RRT outperforms state of the art kinodynamic planners in efficiency, and also provides a shorter path finish time than a steering function free method. The learned local planner policy and accompanying reachability estimator demonstrate transferability to the previously unseen experimental environments, making RL-RRT fast because the expensive computations are replaced with simple neural network inference. Video: https://youtu.be/dDMVMTOI8KY <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1907.04799v2-abstract-full').style.display = 'none'; document.getElementById('1907.04799v2-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 July, 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">Accepted to Robotics and Automation Letters in June 2019</span> </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Journal ref:</span> Robotics and Automation Letters 2019 </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1902.09458">arXiv:1902.09458</a> <span> [<a href="https://arxiv.org/pdf/1902.09458">pdf</a>, <a href="https://arxiv.org/format/1902.09458">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"> Long-Range Indoor Navigation with PRM-RL </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Francis%2C+A">Anthony Francis</a>, <a href="/search/cs?searchtype=author&query=Faust%2C+A">Aleksandra Faust</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+H+L">Hao-Tien Lewis Chiang</a>, <a href="/search/cs?searchtype=author&query=Hsu%2C+J">Jasmine Hsu</a>, <a href="/search/cs?searchtype=author&query=Kew%2C+J+C">J. Chase Kew</a>, <a href="/search/cs?searchtype=author&query=Fiser%2C+M">Marek Fiser</a>, <a href="/search/cs?searchtype=author&query=Lee%2C+T+E">Tsang-Wei Edward Lee</a> </p> <p class="abstract mathjax"> <span class="has-text-black-bis has-text-weight-semibold">Abstract</span>: <span class="abstract-short has-text-grey-dark mathjax" id="1902.09458v2-abstract-short" style="display: inline;"> Long-range indoor navigation requires guiding robots with noisy sensors and controls through cluttered environments along paths that span a variety of buildings. We achieve this with PRM-RL, a hierarchical robot navigation method in which reinforcement learning agents that map noisy sensors to robot controls learn to solve short-range obstacle avoidance tasks, and then sampling-based planners map… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1902.09458v2-abstract-full').style.display = 'inline'; document.getElementById('1902.09458v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1902.09458v2-abstract-full" style="display: none;"> Long-range indoor navigation requires guiding robots with noisy sensors and controls through cluttered environments along paths that span a variety of buildings. We achieve this with PRM-RL, a hierarchical robot navigation method in which reinforcement learning agents that map noisy sensors to robot controls learn to solve short-range obstacle avoidance tasks, and then sampling-based planners map where these agents can reliably navigate in simulation; these roadmaps and agents are then deployed on robots, guiding them along the shortest path where the agents are likely to succeed. Here we use Probabilistic Roadmaps (PRMs) as the sampling-based planner, and AutoRL as the reinforcement learning method in the indoor navigation context. We evaluate the method in simulation for kinematic differential drive and kinodynamic car-like robots in several environments, and on differential-drive robots at three physical sites. Our results show PRM-RL with AutoRL is more successful than several baselines, is robust to noise, and can guide robots over hundreds of meters in the face of noise and obstacles in both simulation and on robots, including over 5.8 kilometers of physical robot navigation. Video: https://youtu.be/xN-OWX5gKvQ <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1902.09458v2-abstract-full').style.display = 'none'; document.getElementById('1902.09458v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 22 February, 2020; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 25 February, 2019; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 2019. </p> <p class="comments is-size-7"> <span class="has-text-black-bis has-text-weight-semibold">Comments:</span> <span class="has-text-grey-dark mathjax">Accepted to T-RO</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1811.12651">arXiv:1811.12651</a> <span> [<a href="https://arxiv.org/pdf/1811.12651">pdf</a>, <a href="https://arxiv.org/format/1811.12651">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"> PEARL: PrEference Appraisal Reinforcement Learning for Motion Planning </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Faust%2C+A">Aleksandra Faust</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+H+L">Hao-Tien Lewis Chiang</a>, <a href="/search/cs?searchtype=author&query=Tapia%2C+L">Lydia Tapia</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="1811.12651v1-abstract-short" style="display: inline;"> Robot motion planning often requires finding trajectories that balance different user intents, or preferences. One of these preferences is usually arrival at the goal, while another might be obstacle avoidance. Here, we formalize these, and similar, tasks as preference balancing tasks (PBTs) on acceleration controlled robots, and propose a motion planning solution, PrEference Appraisal Reinforceme… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1811.12651v1-abstract-full').style.display = 'inline'; document.getElementById('1811.12651v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1811.12651v1-abstract-full" style="display: none;"> Robot motion planning often requires finding trajectories that balance different user intents, or preferences. One of these preferences is usually arrival at the goal, while another might be obstacle avoidance. Here, we formalize these, and similar, tasks as preference balancing tasks (PBTs) on acceleration controlled robots, and propose a motion planning solution, PrEference Appraisal Reinforcement Learning (PEARL). PEARL uses reinforcement learning on a restricted training domain, combined with features engineered from user-given intents. PEARL's planner then generates trajectories in expanded domains for more complex problems. We present an adaptation for rejection of stochastic disturbances and offer in-depth analysis, including task completion conditions and behavior analysis when the conditions do not hold. PEARL is evaluated on five problems, two multi-agent obstacle avoidance tasks and three that stochastically disturb the system at run-time: 1) a multi-agent pursuit problem with 1000 pursuers, 2) robot navigation through 900 moving obstacles, which is is trained with in an environment with only 4 static obstacles, 3) aerial cargo delivery, 4) two robot rendezvous, and 5) flying inverted pendulum. Lastly, we evaluate the method on a physical quadrotor UAV robot with a suspended load influenced by a stochastic disturbance. The video, https://youtu.be/ZkFt1uY6vlw contains the experiments and visualization of the simulations. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1811.12651v1-abstract-full').style.display = 'none'; document.getElementById('1811.12651v1-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 November, 2018; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> November 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">20 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/1809.10124">arXiv:1809.10124</a> <span> [<a href="https://arxiv.org/pdf/1809.10124">pdf</a>, <a href="https://arxiv.org/format/1809.10124">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 Navigation Behaviors End-to-End with AutoRL </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chiang%2C+H+L">Hao-Tien Lewis Chiang</a>, <a href="/search/cs?searchtype=author&query=Faust%2C+A">Aleksandra Faust</a>, <a href="/search/cs?searchtype=author&query=Fiser%2C+M">Marek Fiser</a>, <a href="/search/cs?searchtype=author&query=Francis%2C+A">Anthony Francis</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="1809.10124v2-abstract-short" style="display: inline;"> We learn end-to-end point-to-point and path-following navigation behaviors that avoid moving obstacles. These policies receive noisy lidar observations and output robot linear and angular velocities. The policies are trained in small, static environments with AutoRL, an evolutionary automation layer around Reinforcement Learning (RL) that searches for a deep RL reward and neural network architectu… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1809.10124v2-abstract-full').style.display = 'inline'; document.getElementById('1809.10124v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1809.10124v2-abstract-full" style="display: none;"> We learn end-to-end point-to-point and path-following navigation behaviors that avoid moving obstacles. These policies receive noisy lidar observations and output robot linear and angular velocities. The policies are trained in small, static environments with AutoRL, an evolutionary automation layer around Reinforcement Learning (RL) that searches for a deep RL reward and neural network architecture with large-scale hyper-parameter optimization. AutoRL first finds a reward that maximizes task completion, and then finds a neural network architecture that maximizes the cumulative of the found reward. Empirical evaluations, both in simulation and on-robot, show that AutoRL policies do not suffer from the catastrophic forgetfulness that plagues many other deep reinforcement learning algorithms, generalize to new environments and moving obstacles, are robust to sensor, actuator, and localization noise, and can serve as robust building blocks for larger navigation tasks. Our path-following and point-to-point policies are respectively 23% and 26% more successful than comparison methods across new environments. Video at: https://youtu.be/0UwkjpUEcbI <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1809.10124v2-abstract-full').style.display = 'none'; document.getElementById('1809.10124v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 1 February, 2019; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 26 September, 2018; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 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">Accepted to RA-L/ICRA 2019. Chiang and Faust contributed 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/1805.11597">arXiv:1805.11597</a> <span> [<a href="https://arxiv.org/pdf/1805.11597">pdf</a>, <a href="https://arxiv.org/format/1805.11597">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> <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"> Deep Neural Networks for Swept Volume Prediction Between Configurations </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chiang%2C+H+L">Hao-Tien Lewis Chiang</a>, <a href="/search/cs?searchtype=author&query=Faust%2C+A">Aleksandra Faust</a>, <a href="/search/cs?searchtype=author&query=Tapia%2C+L">Lydia Tapia</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="1805.11597v1-abstract-short" style="display: inline;"> Swept Volume (SV), the volume displaced by an object when it is moving along a trajectory, is considered a useful metric for motion planning. First, SV has been used to identify collisions along a trajectory, because it directly measures the amount of space required for an object to move. Second, in sampling-based motion planning, SV is an ideal distance metric, because it correlates to the likeli… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1805.11597v1-abstract-full').style.display = 'inline'; document.getElementById('1805.11597v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1805.11597v1-abstract-full" style="display: none;"> Swept Volume (SV), the volume displaced by an object when it is moving along a trajectory, is considered a useful metric for motion planning. First, SV has been used to identify collisions along a trajectory, because it directly measures the amount of space required for an object to move. Second, in sampling-based motion planning, SV is an ideal distance metric, because it correlates to the likelihood of success of the expensive local planning step between two sampled configurations. However, in both of these applications, traditional SV algorithms are too computationally expensive for efficient motion planning. In this work, we train Deep Neural Networks (DNNs) to learn the size of SV for specific robot geometries. Results for two robots, a 6 degree of freedom (DOF) rigid body and a 7 DOF fixed-based manipulator, indicate that the network estimations are very close to the true size of SV and is more than 1500 times faster than a state of the art SV estimation algorithm. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1805.11597v1-abstract-full').style.display = 'none'; document.getElementById('1805.11597v1-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 29 May, 2018; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> May 2018. </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1803.07046">arXiv:1803.07046</a> <span> [<a href="https://arxiv.org/pdf/1803.07046">pdf</a>, <a href="https://arxiv.org/format/1803.07046">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Information Theory">cs.IT</span> </div> </div> <p class="title is-5 mathjax"> Time-Domain Multi-Beam Selection and Its Performance Improvement for mmWave Systems </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Hsiao-Lan Chiang</a>, <a href="/search/cs?searchtype=author&query=Rave%2C+W">Wolfgang Rave</a>, <a href="/search/cs?searchtype=author&query=Fettweis%2C+G">Gerhard Fettweis</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="1803.07046v2-abstract-short" style="display: inline;"> Multi-beam selection is one of the crucial technologies in hybrid beamforming systems for frequency-selective fading channels. Addressing the problem in the frequency domain facilitates the procedure of acquiring observations for analog beam selection. However, it is difficult to improve the quality of the contaminated observations at low SNR. To this end, this paper uses an idea that the signific… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1803.07046v2-abstract-full').style.display = 'inline'; document.getElementById('1803.07046v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1803.07046v2-abstract-full" style="display: none;"> Multi-beam selection is one of the crucial technologies in hybrid beamforming systems for frequency-selective fading channels. Addressing the problem in the frequency domain facilitates the procedure of acquiring observations for analog beam selection. However, it is difficult to improve the quality of the contaminated observations at low SNR. To this end, this paper uses an idea that the significant observations are sparse in the time domain to further enhance the quality of signals as well as the beam selection performance. By exploiting properties of channel impulse responses and circular convolutions in the time domain, we can reduce the size of a Toeplitz matrix in deconvolution to generate periodic true values of coupling coefficients plus random noise signals. An arithmetic mean of these signals yields refined observations with minor noise effects and provides more accurate sparse multipath delay information. As a result, only the refined observations associated with the estimated multipath delay indices have to be taken into account for the analog beam selection problem. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1803.07046v2-abstract-full').style.display = 'none'; document.getElementById('1803.07046v2-abstract-short').style.display = 'inline';">△ Less</a> </span> </p> <p class="is-size-7"><span class="has-text-black-bis has-text-weight-semibold">Submitted</span> 29 November, 2018; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 19 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">Accepted by IEEE International Conference on Communications (ICC), Kansas City, MO, USA, 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/1802.06670">arXiv:1802.06670</a> <span> [<a href="https://arxiv.org/pdf/1802.06670">pdf</a>, <a href="https://arxiv.org/format/1802.06670">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Information Theory">cs.IT</span> </div> </div> <p class="title is-5 mathjax"> Frequency-Selective Hybrid Beamforming Based on Implicit CSI for Millimeter Wave Systems </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Hsiao-Lan Chiang</a>, <a href="/search/cs?searchtype=author&query=Rave%2C+W">Wolfgang Rave</a>, <a href="/search/cs?searchtype=author&query=Kadur%2C+T">Tobias Kadur</a>, <a href="/search/cs?searchtype=author&query=Fettweis%2C+G">Gerhard Fettweis</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="1802.06670v3-abstract-short" style="display: inline;"> Hybrid beamforming is a promising concept to achieve high data rate transmission at millimeter waves. To implement it in a transceiver, many references optimally adapt to a high-dimensional multi-antenna channel but more or less ignore the complexity of the channel estimation. Realizing that received coupling coefficients of the channel and pairs of possible analog beamforming vectors can be used… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1802.06670v3-abstract-full').style.display = 'inline'; document.getElementById('1802.06670v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1802.06670v3-abstract-full" style="display: none;"> Hybrid beamforming is a promising concept to achieve high data rate transmission at millimeter waves. To implement it in a transceiver, many references optimally adapt to a high-dimensional multi-antenna channel but more or less ignore the complexity of the channel estimation. Realizing that received coupling coefficients of the channel and pairs of possible analog beamforming vectors can be used for analog beam selection, we further propose a low-complexity scheme that exploits the coupling coefficients to implement hybrid beamforming. Essentially, the coupling coefficients can be regarded as implicit channel state information (CSI), and the estimates of these coupling coefficients yield alternatives of effective channel matrices of much lower dimension. After calculating the Frobenius norm of these effective channel matrices, it turns out that the effective channel having the largest value of the Frobenius norm provides the solution to hybrid beamforming problem. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1802.06670v3-abstract-full').style.display = 'none'; document.getElementById('1802.06670v3-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, 2018; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 19 February, 2018; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> February 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">Accepted by IEEE International Conference on Communications (ICC), Kansas City, MO, USA, 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/1801.10300">arXiv:1801.10300</a> <span> [<a href="https://arxiv.org/pdf/1801.10300">pdf</a>, <a href="https://arxiv.org/format/1801.10300">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"> Netizen-Style Commenting on Fashion Photos: Dataset and Diversity Measures </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Lin%2C+W+H">Wen Hua Lin</a>, <a href="/search/cs?searchtype=author&query=Chen%2C+K">Kuan-Ting Chen</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+H+Y">Hung Yueh Chiang</a>, <a href="/search/cs?searchtype=author&query=Hsu%2C+W">Winston Hsu</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="1801.10300v1-abstract-short" style="display: inline;"> Recently, deep neural network models have achieved promising results in image captioning task. Yet, "vanilla" sentences, only describing shallow appearances (e.g., types, colors), generated by current works are not satisfied netizen style resulting in lacking engagements, contexts, and user intentions. To tackle this problem, we propose Netizen Style Commenting (NSC), to automatically generate cha… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1801.10300v1-abstract-full').style.display = 'inline'; document.getElementById('1801.10300v1-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1801.10300v1-abstract-full" style="display: none;"> Recently, deep neural network models have achieved promising results in image captioning task. Yet, "vanilla" sentences, only describing shallow appearances (e.g., types, colors), generated by current works are not satisfied netizen style resulting in lacking engagements, contexts, and user intentions. To tackle this problem, we propose Netizen Style Commenting (NSC), to automatically generate characteristic comments to a user-contributed fashion photo. We are devoted to modulating the comments in a vivid "netizen" style which reflects the culture in a designated social community and hopes to facilitate more engagement with users. In this work, we design a novel framework that consists of three major components: (1) We construct a large-scale clothing dataset named NetiLook, which contains 300K posts (photos) with 5M comments to discover netizen-style comments. (2) We propose three unique measures to estimate the diversity of comments. (3) We bring diversity by marrying topic models with neural networks to make up the insufficiency of conventional image captioning works. Experimenting over Flickr30k and our NetiLook datasets, we demonstrate our proposed approaches benefit fashion photo commenting and improve image captioning tasks both in accuracy and diversity. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1801.10300v1-abstract-full').style.display = 'none'; document.getElementById('1801.10300v1-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 January, 2018; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> January 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">The Web Conference (WWW) 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/1709.07273">arXiv:1709.07273</a> <span> [<a href="https://arxiv.org/pdf/1709.07273">pdf</a>, <a href="https://arxiv.org/format/1709.07273">other</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Information Theory">cs.IT</span> </div> <div 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/JSTSP.2018.2826142">10.1109/JSTSP.2018.2826142 <i class="fa fa-external-link" aria-hidden="true"></i></a></span> </div> </div> </div> <p class="title is-5 mathjax"> Hybrid Beamforming Based on Implicit Channel State Information for Millimeter Wave Links </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Hsiao-Lan Chiang</a>, <a href="/search/cs?searchtype=author&query=Rave%2C+W">Wolfgang Rave</a>, <a href="/search/cs?searchtype=author&query=Kadur%2C+T">Tobias Kadur</a>, <a href="/search/cs?searchtype=author&query=Fettweis%2C+G">Gerhard Fettweis</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="1709.07273v3-abstract-short" style="display: inline;"> Hybrid beamforming provides a promising solution to achieve high data rate transmission at millimeter waves. Implementing hybrid beamforming at a transceiver based on available channel state information is a common solution. However, many reference methods ignore the complexity of channel estimation for large antenna arrays or subsequent steps, such as the singular value decomposition of a channel… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1709.07273v3-abstract-full').style.display = 'inline'; document.getElementById('1709.07273v3-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1709.07273v3-abstract-full" style="display: none;"> Hybrid beamforming provides a promising solution to achieve high data rate transmission at millimeter waves. Implementing hybrid beamforming at a transceiver based on available channel state information is a common solution. However, many reference methods ignore the complexity of channel estimation for large antenna arrays or subsequent steps, such as the singular value decomposition of a channel matrix. To this end, we present a low-complexity scheme that exploits implicit channel knowledge to facilitate the design of hybrid beamforming for frequency-selective fading channels. The implicit channel knowledge can be interpreted as couplings between all possible pairs of analog beamforming vectors at the transmitter and receiver over the surrounding channel. Instead of calculating mutual information between large antenna arrays, we focus on small-size coupling matrices between beam patterns selected by using appropriate key parameters as performance indicators. This converts the complicated hybrid beamforming problem to a much simpler one: it amounts to collecting different sets of the large-power coupling coefficients to construct multiple alternatives for an effective channel matrix. Then, the set yielding the largest Frobenius norm (or the largest absolute value of the determinant) of the effective channel provides the solution to the hybrid beamforming problem. It turns out that the proposed method does not require information on MIMO channel and can be simply implemented by the received correlated pilot signals that are supposed to be used for channel estimation. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1709.07273v3-abstract-full').style.display = 'none'; document.getElementById('1709.07273v3-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 May, 2018; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 21 September, 2017; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> September 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 by Journal of Selected Topics in Signal Processing (J-STSP)</span> </p> </li> <li class="arxiv-result"> <div class="is-marginless"> <p class="list-title is-inline-block"><a href="https://arxiv.org/abs/1606.07374">arXiv:1606.07374</a> <span> [<a href="https://arxiv.org/pdf/1606.07374">pdf</a>] </span> </p> <div class="tags is-inline-block"> <span class="tag is-small is-link tooltip is-tooltip-top" data-tooltip="Machine Learning">cs.LG</span> </div> </div> <p class="title is-5 mathjax"> Multi-Stage Temporal Difference Learning for 2048-like Games </p> <p class="authors"> <span class="search-hit">Authors:</span> <a href="/search/cs?searchtype=author&query=Yeh%2C+K">Kun-Hao Yeh</a>, <a href="/search/cs?searchtype=author&query=Wu%2C+I">I-Chen Wu</a>, <a href="/search/cs?searchtype=author&query=Hsueh%2C+C">Chu-Hsuan Hsueh</a>, <a href="/search/cs?searchtype=author&query=Chang%2C+C">Chia-Chuan Chang</a>, <a href="/search/cs?searchtype=author&query=Liang%2C+C">Chao-Chin Liang</a>, <a href="/search/cs?searchtype=author&query=Chiang%2C+H">Han Chiang</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="1606.07374v2-abstract-short" style="display: inline;"> Szubert and Jaskowski successfully used temporal difference (TD) learning together with n-tuple networks for playing the game 2048. However, we observed a phenomenon that the programs based on TD learning still hardly reach large tiles. In this paper, we propose multi-stage TD (MS-TD) learning, a kind of hierarchical reinforcement learning method, to effectively improve the performance for the rat… <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1606.07374v2-abstract-full').style.display = 'inline'; document.getElementById('1606.07374v2-abstract-short').style.display = 'none';">▽ More</a> </span> <span class="abstract-full has-text-grey-dark mathjax" id="1606.07374v2-abstract-full" style="display: none;"> Szubert and Jaskowski successfully used temporal difference (TD) learning together with n-tuple networks for playing the game 2048. However, we observed a phenomenon that the programs based on TD learning still hardly reach large tiles. In this paper, we propose multi-stage TD (MS-TD) learning, a kind of hierarchical reinforcement learning method, to effectively improve the performance for the rates of reaching large tiles, which are good metrics to analyze the strength of 2048 programs. Our experiments showed significant improvements over the one without using MS-TD learning. Namely, using 3-ply expectimax search, the program with MS-TD learning reached 32768-tiles with a rate of 18.31%, while the one with TD learning did not reach any. After further tuned, our 2048 program reached 32768-tiles with a rate of 31.75% in 10,000 games, and one among these games even reached a 65536-tile, which is the first ever reaching a 65536-tile to our knowledge. In addition, MS-TD learning method can be easily applied to other 2048-like games, such as Threes. Based on MS-TD learning, our experiments for Threes also demonstrated similar performance improvement, where the program with MS-TD learning reached 6144-tiles with a rate of 7.83%, while the one with TD learning only reached 0.45%. <a class="is-size-7" style="white-space: nowrap;" onclick="document.getElementById('1606.07374v2-abstract-full').style.display = 'none'; document.getElementById('1606.07374v2-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 July, 2016; <span class="has-text-black-bis has-text-weight-semibold">v1</span> submitted 23 June, 2016; <span class="has-text-black-bis has-text-weight-semibold">originally announced</span> June 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">The version has been accepted by TCIAIG (The first version was sent on 23, October, 2015)</span> </p> </li> </ol> <div class="is-hidden-tablet">  <span class="help" style="display: inline-block;"><a href="https://github.com/arXiv/arxiv-search/releases">Search v0.5.6 released 2020-02-24</a>  </span> </div> </div> </main> <footer> <div class="columns is-desktop" role="navigation" aria-label="Secondary">  <div class="column"> <div class="columns"> <div class="column"> <ul class="nav-spaced"> <li><a href="https://info.arxiv.org/about">About</a></li> <li><a href="https://info.arxiv.org/help">Help</a></li> </ul> </div> <div class="column"> <ul class="nav-spaced"> <li> <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512" class="icon filter-black" role="presentation"><title>contact arXiv</title><desc>Click here to contact arXiv</desc><path d="M502.3 190.8c3.9-3.1 9.7-.2 9.7 4.7V400c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V195.6c0-5 5.7-7.8 9.7-4.7 22.4 17.4 52.1 39.5 154.1 113.6 21.1 15.4 56.7 47.8 92.2 47.6 35.7.3 72-32.8 92.3-47.6 102-74.1 131.6-96.3 154-113.7zM256 320c23.2.4 56.6-29.2 73.4-41.4 132.7-96.3 142.8-104.7 173.4-128.7 5.8-4.5 9.2-11.5 9.2-18.9v-19c0-26.5-21.5-48-48-48H48C21.5 64 0 85.5 0 112v19c0 7.4 3.4 14.3 9.2 18.9 30.6 23.9 40.7 32.4 173.4 128.7 16.8 12.2 50.2 41.8 73.4 41.4z"/></svg> <a href="https://info.arxiv.org/help/contact.html"> Contact</a> </li> <li> <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512" class="icon filter-black" role="presentation"><title>subscribe to arXiv mailings</title><desc>Click here to subscribe</desc><path d="M476 3.2L12.5 270.6c-18.1 10.4-15.8 35.6 2.2 43.2L121 358.4l287.3-253.2c5.5-4.9 13.3 2.6 8.6 8.3L176 407v80.5c0 23.6 28.5 32.9 42.5 15.8L282 426l124.6 52.2c14.2 6 30.4-2.9 33-18.2l72-432C515 7.8 493.3-6.8 476 3.2z"/></svg> <a href="https://info.arxiv.org/help/subscribe"> Subscribe</a> </li> </ul> </div> </div> </div>   <div class="column"> <div class="columns"> <div class="column"> <ul class="nav-spaced"> <li><a href="https://info.arxiv.org/help/license/index.html">Copyright</a></li> <li><a href="https://info.arxiv.org/help/policies/privacy_policy.html">Privacy Policy</a></li> </ul> </div> <div class="column sorry-app-links"> <ul class="nav-spaced"> <li><a href="https://info.arxiv.org/help/web_accessibility.html">Web Accessibility Assistance</a></li> <li> <p class="help"> <a class="a11y-main-link" href="https://status.arxiv.org" target="_blank">arXiv Operational Status <svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 256 512" class="icon filter-dark_grey" role="presentation"><path d="M224.3 273l-136 136c-9.4 9.4-24.6 9.4-33.9 0l-22.6-22.6c-9.4-9.4-9.4-24.6 0-33.9l96.4-96.4-96.4-96.4c-9.4-9.4-9.4-24.6 0-33.9L54.3 103c9.4-9.4 24.6-9.4 33.9 0l136 136c9.5 9.4 9.5 24.6.1 34z"/></svg></a><br> Get status notifications via <a class="is-link" href="https://subscribe.sorryapp.com/24846f03/email/new" target="_blank"><svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 512 512" class="icon filter-black" role="presentation"><path d="M502.3 190.8c3.9-3.1 9.7-.2 9.7 4.7V400c0 26.5-21.5 48-48 48H48c-26.5 0-48-21.5-48-48V195.6c0-5 5.7-7.8 9.7-4.7 22.4 17.4 52.1 39.5 154.1 113.6 21.1 15.4 56.7 47.8 92.2 47.6 35.7.3 72-32.8 92.3-47.6 102-74.1 131.6-96.3 154-113.7zM256 320c23.2.4 56.6-29.2 73.4-41.4 132.7-96.3 142.8-104.7 173.4-128.7 5.8-4.5 9.2-11.5 9.2-18.9v-19c0-26.5-21.5-48-48-48H48C21.5 64 0 85.5 0 112v19c0 7.4 3.4 14.3 9.2 18.9 30.6 23.9 40.7 32.4 173.4 128.7 16.8 12.2 50.2 41.8 73.4 41.4z"/></svg>email</a> or <a class="is-link" href="https://subscribe.sorryapp.com/24846f03/slack/new" target="_blank"><svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 448 512" class="icon filter-black" role="presentation"><path d="M94.12 315.1c0 25.9-21.16 47.06-47.06 47.06S0 341 0 315.1c0-25.9 21.16-47.06 47.06-47.06h47.06v47.06zm23.72 0c0-25.9 21.16-47.06 47.06-47.06s47.06 21.16 47.06 47.06v117.84c0 25.9-21.16 47.06-47.06 47.06s-47.06-21.16-47.06-47.06V315.1zm47.06-188.98c-25.9 0-47.06-21.16-47.06-47.06S139 32 164.9 32s47.06 21.16 47.06 47.06v47.06H164.9zm0 23.72c25.9 0 47.06 21.16 47.06 47.06s-21.16 47.06-47.06 47.06H47.06C21.16 243.96 0 222.8 0 196.9s21.16-47.06 47.06-47.06H164.9zm188.98 47.06c0-25.9 21.16-47.06 47.06-47.06 25.9 0 47.06 21.16 47.06 47.06s-21.16 47.06-47.06 47.06h-47.06V196.9zm-23.72 0c0 25.9-21.16 47.06-47.06 47.06-25.9 0-47.06-21.16-47.06-47.06V79.06c0-25.9 21.16-47.06 47.06-47.06 25.9 0 47.06 21.16 47.06 47.06V196.9zM283.1 385.88c25.9 0 47.06 21.16 47.06 47.06 0 25.9-21.16 47.06-47.06 47.06-25.9 0-47.06-21.16-47.06-47.06v-47.06h47.06zm0-23.72c-25.9 0-47.06-21.16-47.06-47.06 0-25.9 21.16-47.06 47.06-47.06h117.84c25.9 0 47.06 21.16 47.06 47.06 0 25.9-21.16 47.06-47.06 47.06H283.1z"/></svg>slack</a> </p> </li> </ul> </div> </div> </div>  </div> </footer> <script src="https://static.arxiv.org/static/base/1.0.0a5/js/member_acknowledgement.js"></script> </body> </html>

CINXE.COM

Search | arXiv e-print repository