<div dir="ltr"><div dir="ltr"><div class="gmail_default" style="font-size:small"><div dir="ltr"><div style="font-family:arial,helvetica,sans-serif;font-size:12.8px"><font style="font-family:arial,sans-serif;font-size:small;color:rgb(0,0,0);vertical-align:inherit"><font style="vertical-align:inherit"><b style="background-color:rgb(255,255,0)">PLEASE NOTE SPECIAL TIME!</b></font></font></div><div style="font-family:arial,helvetica,sans-serif;font-size:12.8px"><font style="font-family:arial,sans-serif;font-size:small;color:rgb(0,0,0);vertical-align:inherit"><font style="vertical-align:inherit"><b><br></b></font></font></div><div style="font-family:arial,helvetica,sans-serif;font-size:12.8px"><font style="font-family:arial,sans-serif;font-size:small;color:rgb(0,0,0);vertical-align:inherit"><font style="vertical-align:inherit"><b>When:</b>    </font></font><font style="font-family:Arial,Helvetica,sans-serif;font-size:small;vertical-align:inherit"><font style="vertical-align:inherit"><font style="color:rgb(0,0,0)">    Thursday</font><span class="gmail_default" style="font-family:arial,sans-serif;color:rgb(0,0,0)">, November 17th</span><font style="color:rgb(0,0,0)"> at</font><b style="color:rgb(0,0,0)"> <span style="background-color:rgb(255,255,0)"><u>12:30</u></span></b><b><u><font face="verdana, sans-serif" color="#000000" style="background-color:rgb(255,255,0)"> p</font></u></b><font face="verdana, sans-serif"><b><u><font color="#000000" style="background-color:rgb(255,255,0)">m CT</font></u><font color="#000000"><span style="background-color:rgb(255,255,0)">   </span></font></b></font></font></font><br></div></div><div><p style="color:rgb(80,0,80);font-variant-numeric:normal;font-variant-east-asian:normal;font-stretch:normal;line-height:normal;margin:0px"><font face="arial, sans-serif" color="#000000"><font style="vertical-align:inherit"><font style="vertical-align:inherit"><b><span style="background-color:rgb(255,255,0)"><br></span></b></font></font></font></p><div class="gmail_default"><font face="arial, sans-serif"><b><font color="#500050">Where:      </font><font color="#000000"> </font></b><font color="#000000">Talk will be given </font><font color="#0000ff" style="font-weight:bold"><u>live, in-person</u></font><font style="color:rgb(80,0,80);font-weight:bold"> </font><font style="color:rgb(80,0,80)">at</font></font></div><p class="MsoNormal" style="margin:0in;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><font face="arial, sans-serif"><font color="#500050">               </font><font color="#000000">    TTIC, 6045 S. Kenwood Avenue</font></font></p><p class="MsoNormal" style="margin:0in;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><font face="arial, sans-serif" color="#000000">                   5th Floor, Room 530<b> </b></font></p><p class="MsoNormal" style="margin:0in;color:rgb(80,0,80);line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><font face="arial, sans-serif"><b><span style="color:black"><br></span></b></font></p><p class="MsoNormal" style="margin:0in 0in 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><b style="color:rgb(60,64,67);font-family:Roboto,Arial,sans-serif;letter-spacing:0.2px;white-space:pre-wrap">Virtually:</b><span style="font-size:14px;color:rgb(60,64,67);font-family:Roboto,Arial,sans-serif;letter-spacing:0.2px;white-space:pre-wrap">    <i>via</i> Panopto </span>(<b><a href="https://uchicago.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=8660ee27-1edf-4423-8442-af49000135f5" target="_blank">livestream</a></b><span style="font-size:14px;color:rgb(60,64,67);font-family:Roboto,Arial,sans-serif;letter-spacing:0.2px;white-space:pre-wrap">)</span><br></p><p class="MsoNormal" style="margin:0in 0in 0.0001pt;color:rgb(80,0,80);line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><font face="arial, sans-serif"><br></font></p><p class="MsoNormal" style="margin:0in 0in 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><font face="arial, sans-serif"><font style="vertical-align:inherit"><font style="vertical-align:inherit"><font style="color:rgb(80,0,80)"><b>Who: </b> </font><font color="#500050" style="color:rgb(80,0,80)">    </font><font color="#000000"><font color="#500050">    </font></font></font></font></font>Jacob Andreas, MIT</p><p class="MsoNormal" style="margin:0in 0in 0.0001pt;color:rgb(80,0,80);line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><br></p><div class="MsoNormal" align="center" style="margin:0in 0in 8pt;font-size:11pt;text-align:center;line-height:15.6933px;font-family:Calibri,sans-serif"><hr size="2" width="100%" align="center"></div><div><p style="color:rgb(60,64,67);letter-spacing:0.2px;white-space:pre-wrap"><font face="arial, sans-serif"><b style="letter-spacing:0.2px">Title:</b><span style="letter-spacing:0.2px">         Toward <span class="gmail-il">Natural</span> <span class="gmail-il">Language</span> Supervision</span>
</font></p><div class="gmail_default"><font face="arial, sans-serif"><b style="color:rgb(60,64,67);letter-spacing:0.2px;white-space:pre-wrap">Abstract: </b><span style="color:rgb(60,64,67);letter-spacing:0.2px;white-space:pre-wrap">In the age of deep networks, "learning" almost invariably means "learning from examples". Image classifiers are trained with large datasets of images, machine translation systems with corpora of translated sentences, and robot policies with rollouts or demonstrations. When human learners acquire new concepts and skills, we often do so with richer supervision, especially in the form of <span class="gmail-il">language</span>---we learn new concepts from exemplars accompanied by descriptions or definitions, and new skills from demonstrations accompanied by instructions. In <span class="gmail-il">natural</span> <span class="gmail-il">language</span> <span class="gmail-il">processing</span>, recent years have seen a number of successful approaches to learning from task definitions and other forms of auxiliary <span class="gmail-il">language</span>-based supervision. But these successes have been largely confined to tasks that also involve <span class="gmail-il">language</span> as an input and an output---what will it take to make <span class="gmail-il">language</span>-based <span class="gmail-il">training</span> useful for the rest of the machine learning ecosystem? In this talk, I'll present two recent applications of <span class="gmail-il">natural</span> <span class="gmail-il">language</span> supervision to tasks outside the traditional domain of <span class="gmail-il">NLP</span>: using <span class="gmail-il">language</span> to guide visuomotor policy learning and inductive program synthesis. In these applications, <span class="gmail-il">natural</span> <span class="gmail-il">language</span> annotations reveal latent compositional structure in the space of programs and plans, helping models discover reusable abstractions for perception and interaction. This kind of compositional structure is present in many tasks beyond policy learning and program synthesis, and I'll conclude with a brief discussion of how these techniques can be applied even more generally.</span><br style="color:rgb(60,64,67);letter-spacing:0.2px;white-space:pre-wrap"><br style="color:rgb(60,64,67);letter-spacing:0.2px;white-space:pre-wrap"><b style="color:rgb(60,64,67);letter-spacing:0.2px;white-space:pre-wrap">Bio: </b><span style="color:rgb(60,64,67);letter-spacing:0.2px;white-space:pre-wrap">Jacob Andreas is the X Consortium Assistant Professor at MIT. His research aims to build intelligent systems that can communicate effectively using <span class="gmail-il">language</span> and learn from human guidance. Jacob earned his Ph.D. from UC Berkeley, his M.Phil. from Cambridge (where he studied as a Churchill scholar) and his B.S. from Columbia. As a researcher at Microsoft Semantic Machines, he founded the <span class="gmail-il">language</span> generation team and helped develop core pieces of the technology that powers conversational interaction in Microsoft Outlook. He has been named a Samsung AI Researcher of the Year and National Academy of Sciences Kavli Fellow, and has received MIT's Kolokotrones teaching award and paper awards at NAACL and ICML.</span><br></font></div><div class="gmail_default"><b style="white-space:pre-wrap"><font face="arial, sans-serif"><br></font></b></div><div class="gmail_default"><b style="white-space:pre-wrap"><font face="arial, sans-serif">Host: <a href="mailto:chenhao@uchicago.edu" target="_blank">Chenhao Tan</a></font></b><br></div></div><div><br></div></div><div><div style="font-size:12.8px"><font face="arial, helvetica, sans-serif">**************************************************************************************************<br></font></div><div><p class="MsoNormal" style="margin-bottom:0.0001pt">The <font color="#666666" face="tahoma, sans-serif"><i><b><span class="gmail-il">NLP</span></b></i></font><font color="#666666" face="tahoma, sans-serif"><i><b> <span class="gmail-il">Seminar</span></b></i></font><font style="font-size:12.8px"> </font>is a joint weekly meetup of researchers interested in <i><span class="gmail-il">Natural</span> <span class="gmail-il">Language</span> <span class="gmail-il">Processing</span></i> across TTIC and UChicago that often hosts external <span class="gmail-il">NLP</span> researchers to talk about their research. The <span class="gmail-il">seminar</span> is typically held on Thursdays at noon.<br></p><p class="MsoNormal" style="margin-bottom:0.0001pt;font-size:12.8px;font-family:arial,sans-serif"><br></p><p class="MsoNormal" style="margin-bottom:0.0001pt;font-size:12.8px;font-family:arial,sans-serif"><br></p><p class="MsoNormal" style="margin-bottom:0.0001pt;font-size:12.8px;font-family:arial,sans-serif"><br></p><p class="MsoNormal" style="margin-bottom:0.0001pt;font-size:12.8px;font-family:arial,sans-serif"><br></p><p class="MsoNormal" style="margin-bottom:0.0001pt;font-size:12.8px;font-family:arial,sans-serif"><br></p></div></div></div><div><div dir="ltr" class="gmail_signature"><div dir="ltr"><div><span style="font-family:arial,helvetica,sans-serif;font-size:x-small">Mary C. Marre</span><br></div><div><div><font face="arial, helvetica, sans-serif" size="1">Faculty Administrative Support</font></div><div><i><font face="arial, helvetica, sans-serif" color="#3d85c6" size="1"><b>Toyota Technological Institute</b></font></i></div><div><i><font face="arial, helvetica, sans-serif" color="#3d85c6" size="1">6045 S. Kenwood Avenue, Rm 517</font></i></div><div><font size="1"><i><font face="arial, helvetica, sans-serif" color="#3d85c6">Chicago, IL  60637</font></i><br></font></div><div><font size="1"><i><font face="arial, helvetica, sans-serif" color="#3d85c6">773-834-1757</font></i></font></div><div><b><i><a href="mailto:mmarre@ttic.edu" target="_blank"><font face="arial, helvetica, sans-serif" size="1">mmarre@ttic.edu</font></a></i></b></div></div></div></div></div><br></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Thu, Nov 10, 2022 at 6:23 PM Mary Marre <<a href="mailto:mmarre@ttic.edu">mmarre@ttic.edu</a>> wrote:<br></div><blockquote class="gmail_quote" style="margin:0px 0px 0px 0.8ex;border-left:1px solid rgb(204,204,204);padding-left:1ex"><div dir="ltr"><div><div dir="ltr" style="font-size:small"><div style="font-family:arial,helvetica,sans-serif;font-size:12.8px"><font style="font-family:arial,sans-serif;font-size:small;color:rgb(0,0,0);vertical-align:inherit"><font style="vertical-align:inherit"><b style="background-color:rgb(255,255,0)">PLEASE NOTE SPECIAL TIME!</b></font></font></div><div style="font-family:arial,helvetica,sans-serif;font-size:12.8px"><font style="font-family:arial,sans-serif;font-size:small;color:rgb(0,0,0);vertical-align:inherit"><font style="vertical-align:inherit"><b><br></b></font></font></div><div style="font-family:arial,helvetica,sans-serif;font-size:12.8px"><font style="font-family:arial,sans-serif;font-size:small;color:rgb(0,0,0);vertical-align:inherit"><font style="vertical-align:inherit"><b>When:</b>    </font></font><font style="font-family:Arial,Helvetica,sans-serif;font-size:small;vertical-align:inherit"><font style="vertical-align:inherit"><font style="color:rgb(0,0,0)">    Thursday</font><span class="gmail_default" style="font-family:arial,sans-serif;color:rgb(0,0,0)">, November 17th</span><font style="color:rgb(0,0,0)"> at</font><b style="color:rgb(0,0,0)"> <span style="background-color:rgb(255,255,0)"><u>12:30</u></span></b><b><u><font face="verdana, sans-serif" color="#000000" style="background-color:rgb(255,255,0)"> p</font></u></b><font face="verdana, sans-serif"><b><u><font color="#000000" style="background-color:rgb(255,255,0)">m CT</font></u><font color="#000000"><span style="background-color:rgb(255,255,0)">   </span></font></b></font></font></font><br></div></div><div><p style="font-size:small;color:rgb(80,0,80);font-variant-numeric:normal;font-variant-east-asian:normal;font-stretch:normal;line-height:normal;margin:0px"><font face="arial, sans-serif" color="#000000"><font style="vertical-align:inherit"><font style="vertical-align:inherit"><b><span style="background-color:rgb(255,255,0)"><br></span></b></font></font></font></p><div style="font-size:small"><font face="arial, sans-serif"><b><font color="#500050">Where:      </font><font color="#000000"> </font></b><font color="#000000">Talk will be given </font><font color="#0000ff" style="font-weight:bold"><u>live, in-person</u></font><font style="color:rgb(80,0,80);font-weight:bold"> </font><font style="color:rgb(80,0,80)">at</font></font></div><p class="MsoNormal" style="font-size:small;margin:0in;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><font face="arial, sans-serif"><font color="#500050">               </font><font color="#000000">    TTIC, 6045 S. Kenwood Avenue</font></font></p><p class="MsoNormal" style="font-size:small;margin:0in;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><font face="arial, sans-serif" color="#000000">                   5th Floor, Room 530<b> </b></font></p><p class="MsoNormal" style="font-size:small;margin:0in;color:rgb(80,0,80);line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><font face="arial, sans-serif"><b><span style="color:black"><br></span></b></font></p><p class="MsoNormal" style="font-size:small;margin:0in 0in 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><b style="color:rgb(60,64,67);font-family:Roboto,Arial,sans-serif;letter-spacing:0.2px;white-space:pre-wrap">Virtually:</b><span style="font-size:14px;color:rgb(60,64,67);font-family:Roboto,Arial,sans-serif;letter-spacing:0.2px;white-space:pre-wrap">    <i>via</i> Panopto </span>(<b><a href="https://uchicago.hosted.panopto.com/Panopto/Pages/Viewer.aspx?id=8660ee27-1edf-4423-8442-af49000135f5" target="_blank">livestream</a></b><span style="font-size:14px;color:rgb(60,64,67);font-family:Roboto,Arial,sans-serif;letter-spacing:0.2px;white-space:pre-wrap">)</span><br></p><p class="MsoNormal" style="font-size:small;margin:0in 0in 0.0001pt;color:rgb(80,0,80);line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><font face="arial, sans-serif"><br></font></p><p class="MsoNormal" style="font-size:small;margin:0in 0in 0.0001pt;line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><font face="arial, sans-serif"><font style="vertical-align:inherit"><font style="vertical-align:inherit"><font style="color:rgb(80,0,80)"><b>Who: </b> </font><font color="#500050" style="color:rgb(80,0,80)">    </font><font color="#000000"><font color="#500050">    </font></font></font></font></font>Jacob Andreas, MIT</p><p class="MsoNormal" style="font-size:small;margin:0in 0in 0.0001pt;color:rgb(80,0,80);line-height:normal;background-image:initial;background-position:initial;background-size:initial;background-repeat:initial;background-origin:initial;background-clip:initial"><br></p><div class="MsoNormal" align="center" style="font-size:11pt;margin:0in 0in 8pt;text-align:center;line-height:15.6933px;font-family:Calibri,sans-serif"><hr size="2" width="100%" align="center"></div><div><p style="color:rgb(60,64,67);letter-spacing:0.2px;white-space:pre-wrap"><font face="arial, sans-serif"><b style="letter-spacing:0.2px">Title:</b><span style="letter-spacing:0.2px">         Toward Natural Language Supervision</span>
</font></p><div><font face="arial, sans-serif"><b style="color:rgb(60,64,67);letter-spacing:0.2px;white-space:pre-wrap">Abstract: </b><span style="color:rgb(60,64,67);letter-spacing:0.2px;white-space:pre-wrap">In the age of deep networks, "learning" almost invariably means "learning from examples". Image classifiers are trained with large datasets of images, machine translation systems with corpora of translated sentences, and robot policies with rollouts or demonstrations. When human learners acquire new concepts and skills, we often do so with richer supervision, especially in the form of language---we learn new concepts from exemplars accompanied by descriptions or definitions, and new skills from demonstrations accompanied by instructions. In natural language processing, recent years have seen a number of successful approaches to learning from task definitions and other forms of auxiliary language-based supervision. But these successes have been largely confined to tasks that also involve language as an input and an output---what will it take to make language-based training useful for the rest of the machine learning ecosystem? In this talk, I'll present two recent applications of natural language supervision to tasks outside the traditional domain of NLP: using language to guide visuomotor policy learning and inductive program synthesis. In these applications, natural language annotations reveal latent compositional structure in the space of programs and plans, helping models discover reusable abstractions for perception and interaction. This kind of compositional structure is present in many tasks beyond policy learning and program synthesis, and I'll conclude with a brief discussion of how these techniques can be applied even more generally.</span><br style="color:rgb(60,64,67);letter-spacing:0.2px;white-space:pre-wrap"><br style="color:rgb(60,64,67);letter-spacing:0.2px;white-space:pre-wrap"><b style="color:rgb(60,64,67);letter-spacing:0.2px;white-space:pre-wrap">Bio: </b><span style="color:rgb(60,64,67);letter-spacing:0.2px;white-space:pre-wrap">Jacob Andreas is the X Consortium Assistant Professor at MIT. His research aims to build intelligent systems that can communicate effectively using language and learn from human guidance. Jacob earned his Ph.D. from UC Berkeley, his M.Phil. from Cambridge (where he studied as a Churchill scholar) and his B.S. from Columbia. As a researcher at Microsoft Semantic Machines, he founded the language generation team and helped develop core pieces of the technology that powers conversational interaction in Microsoft Outlook. He has been named a Samsung AI Researcher of the Year and National Academy of Sciences Kavli Fellow, and has received MIT's Kolokotrones teaching award and paper awards at NAACL and ICML.</span><br></font></div><div><b style="white-space:pre-wrap"><font face="arial, sans-serif"><br></font></b></div><div><b style="white-space:pre-wrap"><font face="arial, sans-serif">Host: <a href="mailto:chenhao@uchicago.edu" target="_blank">Chenhao Tan</a></font></b><br></div></div><div style="font-size:small"><br></div></div><div><div style="font-size:12.8px"><font face="arial, helvetica, sans-serif">**************************************************************************************************<br></font></div><div><p class="MsoNormal" style="margin-bottom:0.0001pt"><span>The <font color="#666666" face="tahoma, sans-serif"><i><b>NLP</b></i></font></span><font><font color="#666666" face="tahoma, sans-serif"><i><b> <span>Seminar</span></b></i></font><font face="times new roman, serif" style="font-family:Arial,Helvetica,sans-serif;font-size:12.8px"> </font></font><span style="font-family:Arial,Helvetica,sans-serif;font-size:small">is a joint weekly meetup of researchers interested in</span><span style="font-family:Arial,Helvetica,sans-serif;font-size:small"> </span><i style="font-family:Arial,Helvetica,sans-serif;font-size:small"><span>Natural</span> <span>Language</span> <span>Processing</span></i><span style="font-family:Arial,Helvetica,sans-serif;font-size:small"> across TTIC and UChicago that often hosts external </span><span style="font-family:Arial,Helvetica,sans-serif;font-size:small">NLP</span><span style="font-family:Arial,Helvetica,sans-serif;font-size:small"> researchers to talk about their research. The </span><span style="font-family:Arial,Helvetica,sans-serif;font-size:small">seminar</span><span style="font-family:Arial,Helvetica,sans-serif;font-size:small"> is typically held on Thursdays at noon.</span><br></p><p class="MsoNormal" style="font-size:12.8px;margin-bottom:0.0001pt;font-family:arial,sans-serif"><br></p><p class="MsoNormal" style="font-size:12.8px;margin-bottom:0.0001pt;font-family:arial,sans-serif"><br></p><p class="MsoNormal" style="font-size:12.8px;margin-bottom:0.0001pt;font-family:arial,sans-serif"><br></p><p class="MsoNormal" style="font-size:12.8px;margin-bottom:0.0001pt;font-family:arial,sans-serif"></p></div></div></div><div><div dir="ltr"><div dir="ltr"><div><span style="font-family:arial,helvetica,sans-serif;font-size:x-small">Mary C. Marre</span><br></div><div><div><font face="arial, helvetica, sans-serif" size="1">Faculty Administrative Support</font></div><div><i><font face="arial, helvetica, sans-serif" color="#3d85c6" size="1"><b>Toyota Technological Institute</b></font></i></div><div><i><font face="arial, helvetica, sans-serif" color="#3d85c6" size="1">6045 S. Kenwood Avenue, Rm 517</font></i></div><div><font size="1"><i><font face="arial, helvetica, sans-serif" color="#3d85c6">Chicago, IL  60637</font></i><br></font></div><div><font size="1"><i><font face="arial, helvetica, sans-serif" color="#3d85c6">773-834-1757</font></i></font></div><div><b><i><a href="mailto:mmarre@ttic.edu" target="_blank"><font face="arial, helvetica, sans-serif" size="1">mmarre@ttic.edu</font></a></i></b></div></div></div></div></div></div>
</blockquote></div></div>