BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VTIMEZONE
TZID:Europe/Vienna
BEGIN:DAYLIGHT
DTSTART:20180325T030000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20181028T020000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260403T220423Z
UID:5b45d6b1dc3b0494089323@ist.ac.at
DTSTART:20180813T110000
DTEND:20180813T130000
DESCRIPTION:Speaker: Roman Krems\nhosted by Misha Lemeshko\nAbstract: Machi
 ne learning is becoming a new tool for physics and chemistry research. Thi
 s talk will discuss how quantum theory of molecular dynamics can benefit f
 rom machine learning. In particular\, it will be argued that combining mac
 hine learning with quantum dynamics calculations allows one to ask new que
 stions and may help solve problems generally considered unfeasible. The ma
 in focus of this presentation will be on the inverse scattering problem in
  chemical reaction dynamics. I will illustrate a machine-learning approach
  that can be used to build global potential energy surfaces (PES) for reac
 tive molecular systems based on feedback from quantum scattering calculati
 ons. The method is designed to correct for the uncertainties of quantum ch
 emistry calculations and yield potentials that reproduce accurately the re
 action probabilities in a wide range of energies. These surfaces are obtai
 ned automatically and do not require manual fitting of the ab initio energ
 ies with analytical functions. The PES are built from a small number of ab
  initio points by an iterative process that incrementally samples the most
  relevant parts of the configuration space. Using the dynamical results of
  previous authors as targets\, we show that such feedback loops produce ac
 curate global PES with 30 ab initio energies for the three-dimensional H +
  H2 -> H2 + H reaction and 290 ab inito energies for the six-dimensional O
 H + H2 -> H2O + H reaction. In the second part of the talk\, I will illust
 rate how machine learning can be used for extrapolation of properties of c
 omplex quantum systems. I will describe a machine-learning method for pred
 icting sharp transitions in a Hamiltonian phase diagram by extrapolation. 
 The method is based on Gaussian Process regression with a combination of k
 ernels chosen through an iterative procedure maximizing the predicting pow
 er of the kernels. The method is capable of extrapolating across the trans
 ition lines. The calculations within a given phase can be used to predict 
 not only the closest sharp transition\, but also a transition removed from
  the available data by a separate phase. This method is thus particularly 
 valuable for searching phase transitions in the parts of the parameter spa
 ce that cannot be probed experimentally or theoretically.
LOCATION:Big Seminar room Ground floor / Office Bldg West (I21.EG.101)\, IS
 TA
ORGANIZER:msoronda@ist.ac.at
SUMMARY:Roman Krems: Applications of machine learning for quantum dynamics
URL:https://talks-calendar.ista.ac.at/events/1326
END:VEVENT
END:VCALENDAR