Tianjun Zhang

<!DOCTYPE html> <html lang="en-US"> <head> <meta charset="UTF-8"> <script> (function(i,s,o,g,r,a,m){i['GoogleAnalyticsObject']=r;i[r]=i[r]||function(){ (i[r].q=i[r].q||[]).push(arguments)},i[r].l=1*new Date();a=s.createElement(o), m=s.getElementsByTagName(o)[0];a.async=1;a.src=g;m.parentNode.insertBefore(a,m) })(window,document,'script','https://www.google-analytics.com/analytics.js','ga'); ga('create', 'UA-88403851-1', 'auto'); ga('send', 'pageview'); </script> <script src="https://ajax.googleapis.com/ajax/libs/jquery/2.2.3/jquery.min.js"></script> <script src="page.js"></script> <link href='https://fonts.googleapis.com/css?family=Open+Sans:400,300' rel='stylesheet' type='text/css'> <title>Tianjun Zhang</title>  <link href="bootstrap.css" rel="stylesheet">  <link rel="stylesheet" href="style.css"> </head> <body> <div id="content"> <div id="title"> <h1>Tianjun Zhang</h1> </div> <br>   <div class="navbar navbar-default" role="navigation"> <div class="container"> <div class="navbar-header"> <button type="button" class="navbar-toggle" data-toggle="collapse" data-target=".navbar-collapse"> <span class="sr-only">Toggle navigation</span> <span class="icon-bar"></span> <span class="icon-bar"></span> <span class="icon-bar"></span> </button>  </div> <div class="navbar-collapse collapse"> <ul class="nav navbar-nav navbar-left"> <li class="active"><a href="#about">Home</a></li> <li><a href="#Research Summary">Research Summary</a></li> <li><a href="#News">News</a></li> <li><a href="#Publications">Publications</a></li> <li><a href="#work">Experience</a></li> <li><a href="#Services">Services</a></li> </ul> </div> </div> </div> <div class="container" style="margin: 0px -10px; width: 760px;"> <img id='profile-img' style="float:left; margin: 20px 20px 20px 0px" src="images/tianjun_profile.png" alt="Tianjun Zhang"> <p>Doctoral student (Ph.D.), EECS Department, UC Berkeley <p>Email: tianjunz [at] berkeley [dot] edu <p> <a href="https://scholar.google.com/citations?user=UE9jz_MAAAAJ&hl=en">Google Scholar</a> <p>I am a fourth-year PhD student in EECS Deparment, UC Berkeley affilicated with the <a href="https://bair.berkeley.edu/">BAIR</a> lab and the <a href="https://rise.cs.berkeley.edu/">RISE</a> lab. I am advised by Prof. <a href="https://people.eecs.berkeley.edu/~jegonzal/">Joseph E. Gonzalez</a>. My research interests generally lie in designing better RL algorithms and their application to robotics.</p> <p> At Berkeley, I am fortunate to work closely with Prof. <a href="https://people.eecs.berkeley.edu/~svlevine/">Sergey Levine</a> and Prof. <a href="https://people.eecs.berkeley.edu/~jiantao/">Jiantao Jiao</a>. I also collaborated closely with Dr. <a href="https://yuandong-tian.com/">Yuandong Tian</a> from Meta Research and Dr. <a href="https://sites.google.com/site/daibohr/">Bo Dai</a> from Google Brain. </p> <hr> </div> <br> <div class="container" id="bio"> <h3>Research Summary</h3> <p> My research has focused on designing better RL algorithms to solve real-world challenges. To be specific, I am working on the following areas: <ul> <li>Data-driven planning: Lap3[<a href="https://arxiv.org/abs/2106.10544">1</a>], C-Planning[<a href="https://arxiv.org/abs/2110.12080">2</a>]</li> <li>Representation Learning for RL: SPEDE [<a href="https://arxiv.org/abs/2111.11485">3</a>] </li> <li>Better Exploration Criterion: BeBold [<a href="https://arxiv.org/abs/2012.08621">4</a>], MADE[<a href="https://arxiv.org/abs/2106.10268">5</a>]</li> </ul> <p><b>Research opportunities</b>: I am actively looking for students to help with research projects both during the semester and over the summer. If you are interested, please send me an email.</p> <hr> </div> <div class="container" id="news"> <h3>News</h3> <p><a2>New!</a2> Two papers were accepted to ICLR 2022: <a href="https://arxiv.org/abs/2110.12080">C-Planning: An Automatic Curriculum for Learning Goal-Reaching Tasks</a> and <a href="https://arxiv.org/abs/2110.03173">Multi-objective Optimization by Learning Space Partitions</a>.</p> <p><a2>New!</a2> Two papers were accepted to NeurIPS 2021 Deep RL workshop: <a href="https://arxiv.org/abs/2110.12080">C-Planning: An Automatic Curriculum for Learning Goal-Reaching Tasks</a> and <a href="https://openreview.net/pdf/45363cfb2c1b1123fc49b52916f8f3a451e09bbd.pdf">Graph Backup: Data Efficient Backup Exploiting Markovian Data</a>. </p> <p><a2>New!</a2> One paper was accepted to EMNLP 2021: <a href="https://arxiv.org/abs/2007.04973">Contrastive Code Representation Learning</a>.</p> <p><a2>New!</a2> Three papers were accepted to NeurIPS 2021: <a href="https://arxiv.org/abs/2012.08621">BeBold: Exploration Beyond the Boundary of Explored Regions</a>, <a href="https://arxiv.org/abs/2106.10268">MADE: Exploration via Maximizing Deviation from Explored Regions</a>, <a href="https://arxiv.org/abs/2106.10544">Learning Space Partitions for Path Planning</a>. </p> <p> One paper were accepted to NeurIPS 2019: <a href="https://arxiv.org/abs/1906.04596">ANODEV2: A Coupled Neural ODE Evolution Framework</a>.</p> <table id="news-table" style="padding: 0px 20px"></table> <hr> </div> <div class="container" id="research"> <h3>Selected Publications</h3> <p style="margin: 20px 0px">See <a href="https://scholar.google.com/citations?user=UE9jz_MAAAAJ&hl=en">Google Scholar</a> for a complete and up-to-date list of publications.</p> <table id="research-table"></table> <p style="margin: 0px 20px">*Equal contribution.</p> <hr> </div> <div class="container"> <p id="copyright">漏 2022 Tianjun Zhang</p> </div>  <div class="spacer" style="clear: both;"></div> </div>  </body> </html>

CINXE.COM

Tianjun Zhang