BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Solving the Many-Electron Schrödinger Equation with Deep Neural N
 etworks  - Matthew Foulkes (Imperial College)
DTSTART:20200910T130000Z
DTEND:20200910T140000Z
UID:TALK150934@talks.cam.ac.uk
CONTACT:Jan Behrends
DESCRIPTION:Exact wavefunctions of interesting many-electron systems are N
 P-hard to compute in general\, but approximations can be found using polyn
 omially-scaling algorithms. The challenge is to find an approximate wavefu
 nction that is simple enough to evaluate and yet has enough variational fr
 eedom to produce accurate results. Neural networks have shown impressive p
 ower as accurate practical function approximators and promise as compact w
 avefunctions for spin systems\, but the Pauli principle complicates networ
 k representations of many-fermion wavefunctions. Here we introduce a fully
  antisymmetric deep learning architecture\, Fermi Net\, able to approximat
 e the wavefunctions of atoms and small molecules to remarkable accuracy. F
 or example\, we predict the dissociation curves of the nitrogen molecule a
 nd the hydrogen chain\, two challenging strongly-correlated systems\, to s
 ignificantly higher precision than the coupled cluster method\, widely con
 sidered the best scalable method for quantum chemistry. This work opens th
 e possibility of accurate direct optimisation of wavefunctions for previou
 sly intractable molecules and solids.
LOCATION:Details of video conferencing will be distributed nearer the time
 .
END:VEVENT
END:VCALENDAR