BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Statistics
SUMMARY:Variational Bayesian inference for PDE based inver
se problems - Ieva Kazlauskaite (University of Cam
bridge)
DTSTART;TZID=Europe/London:20230127T140000
DTEND;TZID=Europe/London:20230127T150000
UID:TALK194893AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/194893
DESCRIPTION:In this talk I will discuss inference in PDE based
Bayesian inverse problems and present our recent
work on variational inference as an alternative to
MCMC for this class of problems. In this work\, w
e propose a family of Gaussian trial distributions
parametrised by precision matrices\, taking advan
tage of the inherent sparsity of the inverse probl
em encoded in its finite element discretisation.
We utilise stochastic optimisation to efficiently
estimate the variational objective and provide an
empirical assessment of the performance. Furthermo
re\, I will mention some recent work that utilises
physics-informed neural network as an alternative
to the classical finite element solvers and illus
trate how these can be used in PDE based forward a
nd inverse problems.
LOCATION:MR12\, Centre for Mathematical Sciences
CONTACT:Qingyuan Zhao
END:VEVENT
END:VCALENDAR