Monte Carlo Inference Via Greedy Importance Sampling
- Dale Schuurmans, AICML
- Aalo Bistritz, Cross Cancer Institute
- Finnegan Southey
We present a new method for conducting Monte Carlo inference in graphical models which com bines explicit search with generalized importance sampling. The idea is to reduce the variance of importance sampling by searching for signifi cant points in the target distribution. We prove that it is possible to introduce search and still maintain unbiasedness. We then demonstrate our procedure on a few simple inference tasks and show that it can improve the inference quality of standard MCMC methods, including Gibbs sam pling, Metropolis sampling, and Hybrid Monte Carlo. This paper extends previous work which showed how greedy importance sampling could be correctly realized in the onedimensional case.
Citation
D. Schuurmans, A. Bistritz, F. Southey. "Monte Carlo Inference Via Greedy Importance Sampling". Conference on Uncertainty in Artificial Intelligence (UAI), July 2000.Keywords: | greedy, importance, machine learning |
Category: | In Conference |
BibTeX
@incollection{Schuurmans+al:UAI00, author = {Dale Schuurmans and Aalo Bistritz and Finnegan Southey}, title = {Monte Carlo Inference Via Greedy Importance Sampling}, booktitle = {Conference on Uncertainty in Artificial Intelligence (UAI)}, year = 2000, }Last Updated: August 13, 2007
Submitted by Russ Greiner