Monte Carlo inference via greedy importance sampling

Monte Carlo inference via greedy importance sampling (2000)

Cached

Download Links

Other Repositories/Bibliography

DBLP

by Dale Schuurmans , Finnegan Southey

Venue:	In Proceedings UAI
Citations:	3 - 1 self

BibTeX

@INPROCEEDINGS{Schuurmans00montecarlo,
    author = {Dale Schuurmans and Finnegan Southey},
    title = {Monte Carlo inference via greedy importance sampling},
    booktitle = {In Proceedings UAI},
    year = {2000},
    pages = {523--532},
    publisher = {Morgan Kaufmann Publishers}
}

OpenURL

Abstract

We present a new method for conducting Monte Carlo inference in graphical models which combines explicit search with generalized importance sampling. The idea is to reduce the variance of importance sampling by searching for significant 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 sampling, Metropolis sampling, and Hybrid Monte Carlo. This paper extends previous work which showed how greedy importance sampling could be correctly realized in the one-dimensional case. 1 Introduction It is well known that general inference and learning with Bayesian networks is computationally hard [DL93, Rot93], and it is therefore necessary to consider restricted architectures [Pea88], or heuristic and approximate algorithms to perform these tasks [JGJS98, Fre9...

Keyphrases

monte carlo inference greedy importance approximate algorithm new method significant point gibbs sampling target distribution importance sampling hybrid monte carlo simple inference task one-dimensional case previous work hard dl93 greedy importance sampling generalized importance sampling metropolis sampling inference quality restricted architecture bayesian network standard mcmc method graphical model general inference explicit search