BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:A transferable active-learning strategy for reactive molecular for
 ce fields   - Fernanda Duarte\, University of Oxford
DTSTART:20210712T153000Z
DTEND:20210712T163000Z
UID:TALK161407@talks.cam.ac.uk
CONTACT:Bingqing Cheng
DESCRIPTION:Predictive molecular simulations require fast\, accurate and r
 eactive interatomic potentials. Machine learning offers a promising approa
 ch to construct such potentials by fitting energies and forces to high-lev
 el quantum-mechanical data\, but doing so typically requires considerable 
 human intervention and data volume.  \n\n \n\nIn this talk\, I will presen
 t our efforts to tackle these challenges by introducing automation and mac
 hine-learned potentials to study reactions mechanisms in the condensed pha
 se\, We have recently shown that by leveraging hierarchical and active lea
 rning\, Gaussian Approximation Potential (GAP) models can be developed for
  diverse chemical systems in an efficient manner\, requiring only hundreds
  to a few thousand energy and gradient evaluations on a reference potentia
 l energy surface. The approach uses separate intra- and inter-molecular fi
 ts and active learning to maximise a prospective error metric used to quan
 tify accuracy. We have applied this strategy to study a range of molecular
  systems: from bulk solvents to chemical reactivity\, including a bifurcat
 ing Diels–Alder reaction in the gas phase and non-equilibrium dynamics (
 SN2 reaction) in explicit solvent. While promising\, many challenges remai
 n which need to be addressed in order to expand the applicability of this 
 strategy to increasingly complex systems.
LOCATION:virtual ZOOM meeting ID: 263 591 6003\, Passcode: 000042\, https:
 //us02web.zoom.us/j/2635916003?pwd=ZlBEQnRENGwxNmJGMENGMWxjak5nUT09
END:VEVENT
END:VCALENDAR