Infinite Mixtures of Trees
- Sergey Kirshner, AICML, Department of Computing Science, University of Alberta
- Padhraic Smyth, University of California Irvine
Finite mixtures of tree-structured distributions have been shown to be efficient and effective in modeling multivariate distributions. Using Dirichlet processes, we extend this approach to allow countably many tree-structured mixture components. The resulting Bayesian framework allows us to deal with the problem of selecting the number of mixture components by computing the posterior distribution over the number of components and integrating out the components by Bayesian model averaging. We apply the proposed framework to identify the number and the properties of predominant precipitation patterns in historical archives of climate data.
Citation
S. Kirshner, P. Smyth. "Infinite Mixtures of Trees". International Conference on Machine Learning (ICML), June 2007.| Keywords: | Dirichlet Processes, Chow-Liu Trees, Non-Parameteric Bayesian Modeling, machine learning |
| Category: | In Conference |
BibTeX
@incollection{Kirshner+Smyth:ICML07,
author = {Sergey Kirshner and Padhraic Smyth},
title = {Infinite Mixtures of Trees},
booktitle = {International Conference on Machine Learning (ICML)},
year = 2007,
}Last Updated: April 19, 2007Submitted by Sergey Kirshner
