BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Bayesian Inference with Kernels - Arthur Gretton\, UCL
DTSTART:20101110T140000Z
DTEND:20101110T150000Z
UID:TALK26857@talks.cam.ac.uk
CONTACT:Dr. Pushmeet Kohli
DESCRIPTION:An embedding of probability distributions into a reproducing k
 ernel Hilbert space (RKHS) has been introduced: like the characteristic fu
 nction\, this provides a unique representation of a probability distributi
 on in a high dimensional feature space. This representation forms the basi
 s of an inference procedure on graphical models\, where the likelihoods ar
 e represented as RKHS functions. The resulting algorithm is completely non
 parametric: all aspects of the model are represented implicitly\, and lear
 ned from a training sample. Both exact inference on trees and loopy BP on 
 pairwise Markov random fields are demonstrated.  \n\nKernel message passin
 g can be applied to general domains where kernels are defined\, handling c
 hallenging cases such as discrete variables with huge domains\, or very co
 mplex\, non-Gaussian continuous distributions. In experiments\, the approa
 ch outperforms state-of-the-art techniques in a cross-lingual document ret
 rieval task and in the prediction of depth from 2-D images. Finally\, time
  permitting\, a more general kernelized Bayes` law will be described\, in 
 which a prior distribution embedding is updated to provide a posterior dis
 tribution embedding. This last approach makes weaker assumptions on the un
 derlying distributions\, but is somewhat more complex to implement.\n\nJoi
 nt work with Danny Bickson\, Kenji Fukumizu\, Carlos Guestrin\, Yucheng Lo
 w\, Le Song\n
LOCATION:Small public lecture room\, Microsoft Research Ltd\, 7 J J Thomso
 n Avenue (Off Madingley Road)\, Cambridge
END:VEVENT
END:VCALENDAR