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:Generalised Particle Filters with Gaussian Mixture
s - Li\, K (Uppsala University)
DTSTART;TZID=Europe/London:20140425T115000
DTEND;TZID=Europe/London:20140425T122500
UID:TALK52186AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/52186
DESCRIPTION:Stochastic filtering is defined as the estimation
of a partially observed dynamical system.\nA massi
ve scientific and computational effort has been de
dicated to the development of numerical methods fo
r approximating the solution of the filtering prob
lem. Approximating with Gaussian mixtures has been
very popular since the 1970s\, however the existi
ng work is only based on the success of the numeri
cal implementation and is not theoretically justif
ied.\n\nWe fill this gap and conduct a rigorous an
alysis of a new Gaussian mixture approximation\nto
the solution of the filtering problem. In particu
lar\, we construct the corresponding approximating
algorithm\, deduce the L2-convergence rate and pr
ove a central limit type theorem for the approxima
ting system. In addition\, we show a numerical exa
mple to illustrate some features of this algorithm
. This is joint work with Dan Crisan (Imperial Col
lege London). \n\nReferences: [1] D. Crisan\, K. L
i\, A central limit type theorem for Gaussian mixt
ure approximations to the nonlinear filtering prob
lem\, ArXiv1401:6592\, (2014).\n\n[2] D. Crisan\,
K. Li\, Generalised particle filters with Gaussian
mixtures\, accepted by\nStochastic Processes and
their Applications\, ArXiv1306:0255\, (2013).\n\n[
3] D. Crisan\, K. Li\, Generalised particle filter
s with Gaussian measures\, Proceedings of\n19th Eu
ropean Signal Processing Conference\, Barcelona\,
Spain\, pp. 659-663\, (2011).\n
LOCATION:Seminar Room 1\, Newton Institute
CONTACT:Mustapha Amrani
END:VEVENT
END:VCALENDAR