BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Composite Gaussian processes for probabilistic PES prediction - Fa
 bio Albertani\, University of Cambridge
DTSTART:20191030T141500Z
DTEND:20191030T143500Z
UID:TALK130435@talks.cam.ac.uk
CONTACT:Lisa Masters
DESCRIPTION:The assignment of spectral lines in both the visible region an
 d the UV region has important applications in the study of astral objects 
 such as stars and exoplanets\, or\, closer to us\, in the study of the mol
 ecular composition of the terrestrial atmosphere. However\, the correct as
 signment of spectral lines cannot solely rely on experimental results\, si
 nce those are not always available\, making theoretical predictions necess
 ary.\n\nTo this end\, we extended the use of machine learning techniques\,
  more specifically of Gaussian regression processes\, in PES prediction to
  the use of composite Gaussian regression processes (c-GP) trained at diff
 erent levels of theory\, with different training sets\, as well as differe
 nt coordinates systems. The PES is then given as a sum of probabilistic pr
 ediction corresponding to dense training sets at low levels of theory and 
 sparser training sets on computationally expensive deterministic (or stoch
 astic) methods.\n\nThe study of the most abundant cation in the Universe (
 and simplest polyatomic cation)\, H3+\, is used to assess the performance 
 of c-GPs.
LOCATION:Department of Chemistry\, Cambridge\, Unilever lecture theatre
END:VEVENT
END:VCALENDAR