BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Memento EPFL//
BEGIN:VEVENT
SUMMARY:"Machine learning in chemistry and beyond" (ChE-650) seminar by Pr
 of. Andrew White (University of Rochester): Making cool stuff with deep le
 arning
DTSTART:20211005T151500
DTEND:20211005T161500
DTSTAMP:20260512T003335Z
UID:a812c22b037f3938ebe3db98dbd088ef8968fb2cf1c967985b632911
CATEGORIES:Conferences - Seminars
DESCRIPTION:Andrew White graduated from Rose-Hulman Institute of Technolog
 y in 2008 with a BS in chemical engineering. While at Rose\, he spent a ye
 ar studying at the Otto-von Guericke Universität and the Max Planck Insti
 tute for Dynamics of Complex Technical Systems in Magdeburg\, Germany. Dr.
  White completed a PhD in chemical engineering at the University of Washin
 gton in 2013. The thesis topic was the creation of non-fouling biomimetic 
 surfaces with computational modeling. Next\, Dr. White worked with Profess
 or Greg Voth at University of Chicago as a Post-doctoral fellow in the Ins
 titute for Biophysical Dynamics from 2013-2014. In Chicago\, he developed 
 new methods for combining simulations and experiments. Dr. White joined th
 e University of Rochester in Chemical Engineering in 2015 and is currently
  an associate professor. He has joint appointments in the Chemistry Depart
 ment\, Biophysics\, Materials Science\, and Data Science programs. Dr. Whi
 te received a National Science Foundation CAREER award in 2018 and an Outs
 tanding Young Investigator Award from the National Institutes of Health in
  2020. Dr. White has authored a textbook on deep learning for molecules an
 d materials\, which is freely available at https://whitead.github.io/dmol-
 book.\nDeep learning has begun a renaissance in chemistry and materials. W
 e can devise and fit models to predict molecular properties in a few hours
  and deploy them in a web browser. We can create novel generative models t
 hat were previously PhD theses in an afternoon. In my group\, we’re expl
 oring deep learning in soft materials and molecules. We are focused on two
  major problems: interpretability and data scarcity. Now that we can make 
 deep learning models to predict any molecular property ad naseum\, what ca
 n we learn? I will discuss our recent efforts on interpreting deep learnin
 g models through symbolic regression and counterfactuals. Data scarcity is
  a common problem in chemistry: how can we learn new properties without si
 gnificant expense of experiments? One method is in judicious choose of exp
 eriments\, which can be done with active learning. Another approach is pre
 -training or meta-leraning\, which tries to exploit related data. I will c
 over recent progress in these areas. Finally\, one consequence of the stat
 e of deep learning is that you can just make cool things in chemistry with
  minimal effort. I’ll review a few fun projects\, including making molec
 ules by banging on the keyboard\, doing math with emojis\, finding chemica
 l entities in HTML\, and doing molecular dynamics with ImageNet derived po
 tentials.\n\nIf you are on the faculty job market Andrew would love to dis
 cuss an opening at the University of Rochester opening and answer question
 s. Please contact the organizer (Kevin Jablonka) or Andrew to schedule a d
 iscussion.
LOCATION:https://epfl.zoom.us/j/64473017589?pwd=Vmpnd1pleGhEb1hFb3kxUlNIUW
 JyQT09
STATUS:CONFIRMED
END:VEVENT
END:VCALENDAR