BEGIN:VCALENDAR
PRODID:-//Columba Systems Ltd//NONSGML CPNG/SpringViewer/ICal Output/3.3-
M3//EN
VERSION:2.0
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
DTSTAMP:20190912T151152Z
DTSTART:20190924T110000Z
DTEND:20190924T120000Z
SUMMARY:Ajay Jasra - Unbiased estimation of the gradient of the log-likel
ihood for partially observed diffusions
UID:{http://www.columbasystems.com/customers/uom/gpp/eventid/}bf5-k087pqk
n-bqty1s
DESCRIPTION:Join us for this research seminar\, part of the SQUIDS (Stati
stics\, quantification of uncertainty\, inverse problems and data scienc
e) seminar series.\n\nAbstract: In this work we present a new coupled si
mulation method for the unbiased estimation of the gradient of the log-l
ikelihood for a class of partially observed diffusion processes. This is
of interest in stochastic gradient algorithms\, which typically require
such unbiasedness. Intrinsically\, given only access to standard time-d
iscretizations of diffusion processes (such as Euler)\, we present a met
hodology which provides unbiased estimates of gradient of the log-likeli
hood which has no time-discretization error. This method also provides e
stimates with finite variance and expected cost. Some preliminary simula
tion results are also given. This is a joint work with Jeremy Heng (ESSE
C Singapore) and Jeremie Houssineau (Warwick).
STATUS:TENTATIVE
TRANSP:TRANSPARENT
CLASS:PUBLIC
LOCATION:Frank Adams 1\, Alan Turing Building\, Manchester
END:VEVENT
END:VCALENDAR