<html>
<head>
<meta http-equiv="Content-Type" content="text/html; charset=us-ascii">
</head>
<body style="word-wrap: break-word; -webkit-nbsp-mode: space; line-break: after-white-space;" class="">
<div class="" style="font-family: AvenirNext-Regular;"><span class="">Dear all,</span></div>
<span class="" style="font-family: AvenirNext-Regular;">
<div class=""><span class=""><br class="">
</span></div>
Tomorrow we will have a Theory Seminar starting at <b class="">1:00PM</b>.</span><span class="" style="font-family: AvenirNext-Regular;"></span><span class="" style="font-family: AvenirNext-Regular;"></span>
<div class="" style="font-family: AvenirNext-Regular;"><br class="">
</div>
<div class=""><font face="AvenirNext-Regular" class="" style="font-family: AvenirNext-Regular;"><span class=""></span></font>
<div class="" style="font-family: AvenirNext-Regular;">Please see below for the details:</div>
<div class="" style="font-family: AvenirNext-Regular;"><br class="">
</div>
<div class="" style="font-family: AvenirNext-Regular;">_____________________________________________________________</div>
<div class="" style="font-family: AvenirNext-Regular;"><b class="">Bluejeans connection:</b> <a href="https://bluejeans.com/801786278" class="">https://bluejeans.com/801786278</a></div>
<div class="" style="font-family: AvenirNext-Regular;"><b class="">Date and time: </b>Monday November 30th, <b class="">1:00 PM</b> <b class="">EST</b></div>
<div class=""><span class="" style="font-family: AvenirNext-Regular;"><b class="">Speaker: </b></span><span class=""><font face="AvenirNext-Regular" class="">Gurtej Kanwar (MIT) </font></span></div>
<div class="" style="font-family: AvenirNext-Regular;"><b class="">Title: </b><font face="AvenirNext-Regular" class="">Ensemble generation for lattice QFT using machine learning.</font></div>
<div class=""><b class="" style="font-family: AvenirNext-Regular;">Abstract: </b><font face="AvenirNext-Regular" class="">Monte Carlo sampling is a powerful method to compute observables in quantum field theories regularized on a discrete spacetime lattice
(LQFT), which is necessary for example to study the non-perturbative behavior of QCD in the low-energy regime. The cost of drawing independent samples is a major bottleneck in such studies. I discuss our recent work demonstrating that generative machine-learning
models can be used to perform Monte Carlo sampling and produce unbiased estimates of observables in LQFT. This work lays out a framework for exactly encoding translational and gauge symmetries in these models, making training practically viable.</font></div>
<div class="" style="font-family: AvenirNext-Regular;"><font face="AvenirNext-Regular" class="">_____________________________________________________________</font></div>
<font face="AvenirNext-Regular" class="" style="font-family: AvenirNext-Regular;"><span class=""></span></font>
<div class="" style="font-family: AvenirNext-Regular;"><span class=""><br class="">
</span></div>
<div class="" style="font-family: AvenirNext-Regular;"><span class="">See you all there!</span></div>
<div class="" style="font-family: AvenirNext-Regular;"><span class=""><br class="">
Best regards,<br class="">
Astrid, Christos, Filippo</span></div>
</div>
</body>
</html>