BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Bayesian Nonparametric Mixture Models - Jurgen Van Gael\, Engineer
 ing department\, University of Cambridge
DTSTART:20091028T163000Z
DTEND:20091028T180000Z
UID:TALK20909@talks.cam.ac.uk
CONTACT:Richard Samworth
DESCRIPTION:Although Bayesian nonparametric models have been around for a 
 while\, recent advances in theory and computational methods have led to ex
 citing new applications of this family of techniques. As a starting point 
 into Bayesian nonparametrics\, we will look at Bayesian nonparametric mixt
 ure models: why they are used and how they are used.\n\nThe Dirichlet proc
 ess will be the basic building block for the mixture models which we discu
 ss. We will discuss different - equivalent - representations of the Dirich
 let process and explain how we can build a mixture model using them. We wi
 ll touch on how to do inference in these models and show some example appl
 ications.\n\nThe discussion will not be going through one paper in particu
 lar but a very readable paper that touches on many issues which we will di
 scuss is: "Bayesian Density Estimation and Inference Using Mixtures":http:
 //www.questia.com/googleScholar.qst?docId=5002233859\, Michael D. Escobar\
 , Mike West\; Journal of the American Statistical Association\, Vol. 90\, 
 1995\n
LOCATION:MR5\, CMS
END:VEVENT
END:VCALENDAR