A permutation-augmented sampler for DP mixture models

Loading...
Thumbnail Image
Penn collection
Departmental Papers (CIS)
Degree type
Discipline
Subject
Computer Sciences
Funder
Grant number
License
Copyright date
Distributor
Related resources
Author
Liang, Percy
Jordan, Michael
Contributor
Abstract

We introduce a new inference algorithm for Dirichlet process mixture models. While Gibbs sampling and variational methods focus on local moves, the new algorithm makes more global moves. This is done by introducing a permutation of the data points as an auxiliary variable. The algorithm is a blocked sampler which alternates between sampling the clustering and sampling the permutation. The key to the efficiency of this approach is that it is possible to use dynamic programming to consider all exponentially many clusterings consistent with a given permutation. We also show that random projections can be used to effectively sample the permutation. The result is a stochastic hill-climbing algorithm that yields burn-in times significantly smaller than those of collapsed Gibbs sampling.

Advisor
Date of presentation
2007-06-01
Conference name
Departmental Papers (CIS)
Conference dates
2023-05-17T07:09:34.000
Conference location
Date Range for Data Collection (Start Date)
Date Range for Data Collection (End Date)
Digital Object Identifier
Series name and number
Volume number
Issue number
Publisher
Publisher DOI
Journal Issue
Comments
Appearing in Proceedings of the 24 th International Conference on Machine Learning, Corvallis, OR, 2007. Copyright 2007 by the author(s)/owner(s).
Recommended citation
Collection