BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Modeling with Bounded Partition Functions - Ryan Prescott Adams (I
 nference Group\, Cavendish Laboratory)
DTSTART:20080716T130000Z
DTEND:20080716T140000Z
UID:TALK12223@talks.cam.ac.uk
CONTACT:Carl Scheffler
DESCRIPTION:Many probabilistic models for data are well expressed using en
 ergy functions.  Typically the (negative) energy is pushed through the exp
 onential function to find the probability distribution over the data.  Inf
 erence in such models is frequently difficult\, as the partition function 
 involves an intractable sum or integral.  I will talk about a trick that I
  used in the Gaussian process density sampler to help sidestep this proble
 m\, and talk about how it could be generalised to other energy-based proba
 bilistic models.  This trick doesn't necessarily make things easier - it j
 ust changes which aspects of the inference problem are difficult.  Nonethe
 less\, I hope it will foster interesting discussion.
LOCATION:TCM Seminar Room\, Cavendish Laboratory\, Department of Physics
END:VEVENT
END:VCALENDAR