BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Quantum Machine Learning in Chemical Space - Felix Faber\, Univers
 ity of Basel 
DTSTART:20190515T103000Z
DTEND:20190515T113000Z
UID:TALK125188@talks.cam.ac.uk
CONTACT:Nick Woods
DESCRIPTION:Computer simulations in chemistry\, materials- and nano-scienc
 e generally rely on a trade-off between accuracy and computational speed. 
 Quantum mechanical methods can come close to experimental values\, but the
  computational cost of these methods grows rapidly with system size and co
 mplexity. Empirical methods\, such as force-fields and coarse-grained mode
 ls\, can calculate properties of larger systems at reasonable timescales b
 ut tend to be limited to specific sets of systems. The nascent field of ma
 chine learning (ML) poses a different approach to this speed/accuracy trad
 e-off by learning system properties through inference instead of direct ca
 lculation. The prediction error of a ML model tends to decrease systematic
 ally with the number of compounds used to train the model. Hence\, given e
 nough training data\, a ML model can in principle reach arbitrary predicti
 ve accuracies. I will discuss some of the challenges that are encountered 
 when ML is applied to predict properties throughout chemical space. These 
 challenges include how to represent an atomic environment\, which combinat
 ion of representations and regressors is best suited for a given property\
 , and how response operators can be used to efficiently learn response pro
 perties.
LOCATION:TCM Seminar Room\, Cavendish Laboratory
END:VEVENT
END:VCALENDAR