BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Optimizing sampling and free energy estimation with normalizing fl
 ows - Peter Wirnsberger\, Cambridge
DTSTART:20220207T143000Z
DTEND:20220207T150000Z
UID:TALK167276@talks.cam.ac.uk
CONTACT:Dr Christoph Schran
DESCRIPTION:Recently there has been a surge in using generative models\, m
 ost notably normalizing flows\, to address challenging problems in statist
 ical mechanics either by developing new learned schemes or by addressing s
 hortcomings of existing techniques. Efficient sampling from high-dimension
 al Boltzmann distributions given an energy function and accurate free ener
 gy estimation are two key problems in the field. In this talk we will focu
 s on both topics and present our recent works on targeted free energy esti
 mation\, in which we combine free energy perturbation and Bennett’s acce
 ptance ratio method with normalizing flows to obtain powerful estimators [
 1\, 2]. We illustrate the potential of this approach by estimating solid f
 ree energies and briefly discuss the limitations of our current model.\n\n
 [1] Wirnsberger\, Ballard et al.\, Targeted free energy estimation via lea
 rned mappings\, J. Chem. Phys. 153\, 144112 (2020)\n\n[2] Wirnsberger\, Pa
 pamakarios\, Ibarz et al.\, Normalizing flows for atomic solids\, arXiv:21
 11.0869 (2021)
LOCATION:Venue to be confirmed
END:VEVENT
END:VCALENDAR