BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Data-Efficient Machine Learning with Chemical and Physical Priors 
 - Johannes Margraf\, Fritz-Haber-Institut der MPG
DTSTART:20210419T153000Z
DTEND:20210419T163000Z
UID:TALK158935@talks.cam.ac.uk
CONTACT:Bingqing Cheng
DESCRIPTION:While machine learning is often discussed in a 'big data' cont
 ext\, for many chemical applications the generation of large reference dat
 abases can be prohibitive. I will talk about how the data-efficiency of ma
 chine learning models can be improved by using chemical or physical priors
 . In particular\, I will discuss the role of size-extensivity\, physical b
 aseline models and hybrid approaches that integrate electronic structure t
 heory and machine learning.
LOCATION:virtual ZOOM meeting ID: 263 591 6003\, Passcode: 000042\, https:
 //us02web.zoom.us/j/2635916003?pwd=ZlBEQnRENGwxNmJGMENGMWxjak5nUT09
END:VEVENT
END:VCALENDAR