BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Foundations of Nonparametric Bayesian Methods (Part III) - Peter O
 rbanz (University of Cambridge)
DTSTART:20081028T110000Z
DTEND:20081028T130000Z
UID:TALK14248@talks.cam.ac.uk
CONTACT:Peter Orbanz
DESCRIPTION:This 3-part tutorial will address a machine learning audience\
 , not\nassumed to be familiar with measure theory or the theory stochastic
 \nprocesses. The course is intended to provide (1) an overview of what\nno
 nparametric Bayesian models exist beyond those already used in\nmachine le
 arning\, and (2) a basic understanding of the mathematical\nconstruction o
 f ''process'' models\, both existing ones and new models\non a variety of 
 possible domains.\n\nPart III: Construction of new models\n\nRecent works 
 in machine learning consider the construction of models\non other domains 
 than the simplex\, that is\, models which\nnonparametrically generate obje
 cts other than probability\ndistributions (such as the binary matrices gen
 erated by the Indian\nBuffet Process). We discuss how nonparametric Bayesi
 an models can be\nconstructed on arbitrary domains\, and what limitations 
 we will have to\nexpect for such constructions.\n\nWebpage:\nhttp://mlg.en
 g.cam.ac.uk/porbanz/npb-tutorial.html
LOCATION:Engineering Department\, CBL Room 438
END:VEVENT
END:VCALENDAR