BEGIN:VCALENDAR
VERSION:2.0
PRODID:icalendar-ruby
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VTIMEZONE
TZID:Europe/Vienna
BEGIN:DAYLIGHT
DTSTART:20170326T030000
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
END:DAYLIGHT
BEGIN:STANDARD
DTSTART:20171029T020000
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20240914T194251Z
UID:58761c772d1de668038487@ist.ac.at
DTSTART:20170419T160000
DTEND:20170419T180000
DESCRIPTION:Speaker: Richard Nickl\nhosted by Laszlo ErdÃ¶s\nAbstract: In t
he age of `data science'\, most important inverse problems encountered in
applied mathematics\, for instance when one wants to recover parameter coe
fficients of PDEs/diffusions/jump processes\, are naturally modelled to in
clude statistical noise or random measurement error. Standard numerical me
thods to solve inverse problems are generally not robust to the presence o
f noise\, particularly for non-linear problems. In the last years the Baye
sian approach has been put forward as a general paradigm to solve statisti
cal inverse problems in a generic way\, most notably by Andrew Stuart and
co-authors\, who have devised successful MCMC algorithms to compute poster
ior distributions in such settings. A key mathematical question is what re
covery guarantees can be given for these algorithms and the associated sta
tistical inference ('uncertainty quantification') procedures. We will disc
uss some recent theorems that indeed demonstrate that the Bayes formalism
successfully\, and optimally\, solves many relevant inverse problems\, bas
ed on proving frequentist contraction and Laplace-Bernstein-von Mises resu
lts for relevant Bayesian posterior distributions.\n\nNo prior knowledge o
f mathematical statistics will be required for this talk\n
LOCATION:Seminar room Big Ground floor / Office Bldg West (I21.EG.101)\, IS
TA
ORGANIZER:jdeanton@ist.ac.at
SUMMARY:Richard Nickl: On Bayes solutions of some non-linear statistical in
verse problems
URL:https://talks-calendar.ista.ac.at/events/413
END:VEVENT
END:VCALENDAR