|
Mike Jordan
Electrical Engineering and Computer Sciences, UC Berkeley, USA
Wednesday 14 October 2009
12.00
Seminar Room B10 (Basement)
Alexandra House, 17 Queen Square, London, WC1N 3AR
Completely Random Measures for Bayesian Nonparametrics
Bayesian nonparametric modeling and inference are based on using general stochastic processes as prior distributions. Despite the great generality of this definition, the great majority of the work in Bayesian nonparametrics is based on only two stochastic processes: the Gaussian process and the Dirichlet process. Motivated by the needs of applications, I present a broader approach to Bayesian nonparametrics in which priors are obtained from a class of stochastic processes known as "completely random measures" (Kingman, 1967). In particular I will present models based on the beta process and the Bernoulli process, and will discuss an application of these models to the analysis of motion capture data in computational vision.
[Joint work with Romain Thibaux, Emily Fox and Erik Sudderth]