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:Isaac Newton Institute Seminar Series
SUMMARY:Metastability and Monte Carlo Methods for Multisca
le Problems - Konstantinos Spiliopoulos ()
DTSTART;TZID=Europe/London:20160621T094500
DTEND;TZID=Europe/London:20160621T103000
UID:TALK66530AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/66530
DESCRIPTION:Rare events\, metastability and Monte Carlo method
s for stochastic dynamical systems have been of c
entral scientific interest for many years now. In
this article we focus on rough energy landscapes\
, that are modeled as multiscale stochastic dynam
ical systems perturbed by small noise. Large devia
tions deals with the estimation of rare events. D
epending on the type of interaction of the fast s
cales with the strength of the noise we get differ
ent behavior\, both for the large deviations and
for the corresponding Monte Carlo methods. We desc
ribe how to design asymptotically provably effici
ent importance sampling schemes for the estimatio
n of associated rare event probabilities\, such as
exit probabilities\,hitting probabilities\, hitt
ing times\, and expectations of functionals of in
terest. Standard Monte Carlo methods perform poorl
y in these kind of problems in the small noise li
mit. In the presence of multiple scales one faces
additional difficulties and straightforward adapt
ation of importance sampling schemes for standard
small noise diffusions will not produce efficient
schemes. Theoretical results are supplemented by
numerical simulation studies.
LOCATION:Seminar Room 1\, Newton Institute
CONTACT:INI IT
END:VEVENT
END:VCALENDAR