PASCAL - Pattern Analysis, Statistical Modelling and Computational Learning

Expectation propagation for infinite mixtures
Thomas P Minka and Zoubin Ghahramani
In: NIPS 2003 Workshop on Nonparametric Bayesian Methods and Infinite Models, Dec, 2003, Whistler, Canada.


This note describes a method for approximate inference in infinite models that uses deterministic Expectation Propagation instead of Monte Carlo. For infinite Gaussian mixtures, the algorithm provides cluster parameter estimates, cluster memberships, and model evidence. Model parameters, such as the expected size of the mixture, can be efficiently tuned via EM with EP as the E-step. The same approach can apply other infinite models such as infinite HMMs.

PDF - Requires Adobe Acrobat Reader or other PDF viewer.
EPrint Type:Conference or Workshop Item (Talk)
Project Keyword:Project Keyword UNSPECIFIED
Subjects:Computational, Information-Theoretic Learning with Statistics
Learning/Statistics & Optimisation
Theory & Algorithms
ID Code:807
Deposited By:Zoubin Ghahramani
Deposited On:30 December 2004