BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Efficient MCMC for Continuous Time Discrete State Systems - Vinaya
 k Rao (UCL)
DTSTART:20111123T110000Z
DTEND:20111123T120000Z
UID:TALK34594@talks.cam.ac.uk
CONTACT:Zoubin Ghahramani
DESCRIPTION:A variety of phenomena are best described using dynamical mode
 ls which operate on a discrete state space and in continuous time. Example
 s include Markov jump processes\, continuous time Bayesian networks\,\nren
 ewal processes and other point processes\, with applications ranging from 
 systems biology\, genetics\, computing networks and human-computer interac
 tions. Posterior computations typically involve approximations\nlike time 
 discretization and can be computationally intensive. In this talk I will d
 escribe recent work on a class of Markov chain Monte Carlo methods that al
 low efficient computations while still being exact. The core idea is to us
 e an auxiliary variable Gibbs sampler based on uniformization\, a represen
 tation of a continuous time\ndynamical system as a Markov chain operating 
 over a discrete set of points drawn from a Poisson process. This is joint 
 work with Yee Whye Teh. If time permits\, I shall also talk about some rec
 ent work on spatial point processes with David Dunson.
LOCATION:Engineering Department\, CBL Room 438
END:VEVENT
END:VCALENDAR