<div dir="ltr"><div dir="ltr"><div class="gmail_default" style="font-size:small"><div class="gmail_default"><div class="gmail_default"><p style="font-variant-numeric:normal;font-variant-east-asian:normal;font-stretch:normal;line-height:normal;margin:0px"><font face="arial, sans-serif"><font style="vertical-align:inherit"><font style="vertical-align:inherit"><b>When:</b>    </font></font><font style="vertical-align:inherit"><font style="vertical-align:inherit">  Monday, December 14th at 11:10 am</font></font><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></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"><b>Where:</b>     </font></font><font color="#000000">Zoom Virtual Talk (</font><b><font color="#0000ff"><a href="https://uchicagogroup.zoom.us/webinar/register/WN_4Spz5uMSTReAa-kCJ8sUlQ" target="_blank">register in advance here</a></font></b><font color="#000000">)</font></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></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"><b>Who: </b>       </font></font>Andrew Gordon Wilson, NYU</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"><span style="color:rgba(0,0,0,0.9)"><b><font face="arial, sans-serif"><br></font></b></span></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"><span style="color:rgba(0,0,0,0.9)"><b><font face="arial, sans-serif"><br></font></b></span></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"><span style="color:rgba(0,0,0,0.9)"><b>Title:</b>        </span>How Do We Build Models That Learn and Generalize?</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"><span style="color:rgba(0,0,0,0.9)"><font face="arial, sans-serif"><br></font></span></p><p class="MsoNormal" style="margin:0in;line-height:normal"><font face="arial, sans-serif"><b><span style="color:rgb(60,64,67);letter-spacing:0.15pt">Abstract: </span></b>To answer scientific questions, and reason about data, we must build models and perform inference within those models.  But how should we approach model construction and inference to make the most successful predictions?  How do we represent uncertainty and prior knowledge?  How flexible should our models be?  Should we use a single model, or multiple different models?  Should we follow a different procedure depending on how much data are available? How do we learn desirable constraints, such as rotation, translation, or reflection symmetries, when they don't improve standard training loss?</font></p><p class="MsoNormal" style="margin:0in;line-height:normal"><font face="arial, sans-serif"><br></font></p><p class="MsoNormal" style="margin:0in;line-height:normal"><font face="arial, sans-serif">In this talk I will present a philosophy for model construction, grounded in probability theory.  I will exemplify this approach with methods that exploit loss surface geometry for scalable and practical Bayesian deep learning, and resolutions to seemingly mysterious generalization behaviour such as double descent. I will also consider prior specification, generalized Bayesian inference, and automatic symmetry learning. The talk will primarily be based on <a href="https://arxiv.org/abs/2002.08791" target="_blank">https://arxiv.org/abs/2002.08791</a>, and it will also touch on <a href="https://arxiv.org/abs/2002.12880" target="_blank">https://arxiv.org/abs/2002.12880</a> and <a href="https://arxiv.org/abs/2010.11882" target="_blank">https://arxiv.org/abs/2010.11882</a>.<br></font></p><p class="MsoNormal" style="margin:0in;line-height:normal"><font face="arial, sans-serif"><br></font></p><p class="MsoNormal" style="margin:0in;line-height:normal"><font face="arial, sans-serif"><b>Bio:</b> Andrew Gordon Wilson is faculty in the Courant Institute of Mathematical Sciences and Center for Data Science at NYU. Before joining NYU, he was an assistant professor at Cornell University from 2016-2019. He was a research fellow in the Machine Learning Department at Carnegie Mellon University from 2014-2016, and completed his PhD at the University of Cambridge in 2014. Andrew's interests include probabilistic modelling, Gaussian processes, Bayesian statistics, physics inspired machine learning, and loss surfaces and generalization in deep learning. His webpage is <a href="https://cims.nyu.edu/~andrewgw" target="_blank">https://cims.nyu.edu/~andrewgw</a>.<br></font></p><div><br></div><div dir="auto"><div style="margin:0px;padding:0px 0px 20px;width:1120px"><div><div id="gmail-m_-1916303418504908038m_171242323807179252gmail-m_-8743081643725194343gmail-:369" style="direction:ltr;margin:8px 0px 0px;padding:0px"><div id="gmail-m_-1916303418504908038m_171242323807179252gmail-m_-8743081643725194343gmail-:1ro" style="overflow:hidden;font-variant-numeric:normal;font-variant-east-asian:normal;font-stretch:normal;line-height:1.5"><div dir="auto"><div><div dir="auto"><b style="font-family:arial,sans-serif">Host:  </b><span style="font-family:arial,sans-serif"><a href="mailto:dougal@ttic.edu" target="_blank"><b>Dougal Sutherland</b></a></span></div></div></div></div></div></div></div></div></div><span style="font-family:arial,sans-serif">For more information on the </span><span style="font-family:arial,sans-serif">colloquium</span><span style="font-family:arial,sans-serif"> series or to subscribe to the mailing list, please see </span><a href="http://www.ttic.edu/colloquium.php" target="_blank" style="font-family:arial,sans-serif">http://www.ttic.edu/colloquium.php</a><span style="font-family:arial,sans-serif">   </span> </div><div class="gmail_default"><br></div><div class="gmail_default"><br></div></div><div><div dir="ltr" class="gmail_signature" data-smartmail="gmail_signature"><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><font face="arial, helvetica, sans-serif">Mary C. Marre</font><div><font face="arial, helvetica, sans-serif">Faculty Administrative Support</font></div><div><i><font face="arial, helvetica, sans-serif" color="#3d85c6"><b>Toyota Technological Institute</b></font></i></div><div><i><font face="arial, helvetica, sans-serif" color="#3d85c6">6045 S. Kenwood Avenue</font></i></div><div><i><font face="arial, helvetica, sans-serif" color="#3d85c6">Room 517</font></i></div><div><i><font face="arial, helvetica, sans-serif" color="#3d85c6">Chicago, IL  60637</font></i></div><div><i><font face="arial, helvetica, sans-serif">p:(773) 834-1757</font></i></div><div><i><font face="arial, helvetica, sans-serif">f: (773) 357-6970</font></i></div><div><b><i><a href="mailto:mmarre@ttic.edu" target="_blank"><font face="arial, helvetica, sans-serif">mmarre@ttic.edu</font></a></i></b></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div><br></div><br><div class="gmail_quote"><div dir="ltr" class="gmail_attr">On Mon, Dec 7, 2020 at 9:25 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><p style="font-variant-numeric:normal;font-variant-east-asian:normal;font-stretch:normal;line-height:normal;margin:0px"><font face="arial, sans-serif"><font style="vertical-align:inherit"><font style="vertical-align:inherit"><b>When:</b>    </font></font><font style="vertical-align:inherit"><font style="vertical-align:inherit">  Monday, December 14th at 11:10 am</font></font><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></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"><b>Where:</b>     </font></font><font color="#000000">Zoom Virtual Talk (</font><b><font color="#0000ff"><a href="https://uchicagogroup.zoom.us/webinar/register/WN_4Spz5uMSTReAa-kCJ8sUlQ" target="_blank">register in advance here</a></font></b><font color="#000000">)</font></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></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"><b>Who: </b>       </font></font>Andrew Gordon Wilson, NYU</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"><span style="color:rgba(0,0,0,0.9)"><b><font face="arial, sans-serif"><br></font></b></span></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"><span style="color:rgba(0,0,0,0.9)"><b><font face="arial, sans-serif"><br></font></b></span></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"><span style="color:rgba(0,0,0,0.9)"><b>Title:</b>        </span>How Do We Build Models That Learn and Generalize?</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"><span style="color:rgba(0,0,0,0.9)"><font face="arial, sans-serif"><br></font></span></p><p class="MsoNormal" style="margin:0in;line-height:normal"><font face="arial, sans-serif"><b><span style="color:rgb(60,64,67);letter-spacing:0.15pt">Abstract: </span></b>To answer scientific questions, and reason about data, we must build models and perform inference within those models.  But how should we approach model construction and inference to make the most successful predictions?  How do we represent uncertainty and prior knowledge?  How flexible should our models be?  Should we use a single model, or multiple different models?  Should we follow a different procedure depending on how much data are available? How do we learn desirable constraints, such as rotation, translation, or reflection symmetries, when they don't improve standard training loss?</font></p><p class="MsoNormal" style="margin:0in;line-height:normal"><font face="arial, sans-serif"><br></font></p><p class="MsoNormal" style="margin:0in;line-height:normal"><font face="arial, sans-serif">In this talk I will present a philosophy for model construction, grounded in probability theory.  I will exemplify this approach with methods that exploit loss surface geometry for scalable and practical Bayesian deep learning, and resolutions to seemingly mysterious generalization behaviour such as double descent. I will also consider prior specification, generalized Bayesian inference, and automatic symmetry learning. The talk will primarily be based on <a href="https://arxiv.org/abs/2002.08791" target="_blank">https://arxiv.org/abs/2002.08791</a>, and it will also touch on <a href="https://arxiv.org/abs/2002.12880" target="_blank">https://arxiv.org/abs/2002.12880</a> and <a href="https://arxiv.org/abs/2010.11882" target="_blank">https://arxiv.org/abs/2010.11882</a>.<br></font></p><p class="MsoNormal" style="margin:0in;line-height:normal"><font face="arial, sans-serif"><br></font></p><p class="MsoNormal" style="margin:0in;line-height:normal"><font face="arial, sans-serif"><b>Bio:</b> Andrew Gordon Wilson is faculty in the Courant Institute of Mathematical Sciences and Center for Data Science at NYU. Before joining NYU, he was an assistant professor at Cornell University from 2016-2019. He was a research fellow in the Machine Learning Department at Carnegie Mellon University from 2014-2016, and completed his PhD at the University of Cambridge in 2014. Andrew's interests include probabilistic modelling, Gaussian processes, Bayesian statistics, physics inspired machine learning, and loss surfaces and generalization in deep learning. His webpage is <a href="https://cims.nyu.edu/~andrewgw" target="_blank">https://cims.nyu.edu/~andrewgw</a>.<br></font></p><div style="font-size:small"><br></div><div dir="auto" style="font-size:small"><div style="margin:0px;padding:0px 0px 20px;width:1120px"><div><div id="gmail-m_-1916303418504908038m_171242323807179252gmail-m_-8743081643725194343gmail-:369" style="direction:ltr;margin:8px 0px 0px;padding:0px"><div id="gmail-m_-1916303418504908038m_171242323807179252gmail-m_-8743081643725194343gmail-:1ro" style="overflow:hidden;font-variant-numeric:normal;font-variant-east-asian:normal;font-stretch:normal;line-height:1.5"><div dir="auto"><div><div dir="auto"><b style="font-family:arial,sans-serif">Host:  </b><span style="font-family:arial,sans-serif"><a href="mailto:dougal@ttic.edu" target="_blank"><b>Dougal Sutherland</b></a></span></div></div></div></div></div></div></div></div></div><span style="font-size:small;font-family:arial,sans-serif">For more information on the </span><span style="font-size:small;font-family:arial,sans-serif">colloquium</span><span style="font-size:small;font-family:arial,sans-serif"> series or to subscribe to the mailing list, please see </span><a href="http://www.ttic.edu/colloquium.php" style="font-size:small;font-family:arial,sans-serif" target="_blank">http://www.ttic.edu/colloquium.php</a><span style="font-size:small;font-family:arial,sans-serif">   </span> </div><div style="font-size:small"><br></div><div style="font-size:small"> <br></div><div><div dir="ltr"><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><div><div dir="ltr"><font face="arial, helvetica, sans-serif">Mary C. Marre</font><div><font face="arial, helvetica, sans-serif">Faculty Administrative Support</font></div><div><i><font face="arial, helvetica, sans-serif" color="#3d85c6"><b>Toyota Technological Institute</b></font></i></div><div><i><font face="arial, helvetica, sans-serif" color="#3d85c6">6045 S. Kenwood Avenue</font></i></div><div><i><font face="arial, helvetica, sans-serif" color="#3d85c6">Room 517</font></i></div><div><i><font face="arial, helvetica, sans-serif" color="#3d85c6">Chicago, IL  60637</font></i></div><div><i><font face="arial, helvetica, sans-serif">p:(773) 834-1757</font></i></div><div><i><font face="arial, helvetica, sans-serif">f: (773) 357-6970</font></i></div><div><b><i><a href="mailto:mmarre@ttic.edu" target="_blank"><font face="arial, helvetica, sans-serif">mmarre@ttic.edu</font></a></i></b></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div></div>
</blockquote></div></div>