AND/OR Importance Sampling
Artificial Intelligence
2012-06-18 v1
Abstract
The paper introduces AND/OR importance sampling for probabilistic graphical models. In contrast to importance sampling, AND/OR importance sampling caches samples in the AND/OR space and then extracts a new sample mean from the stored samples. We prove that AND/OR importance sampling may have lower variance than importance sampling; thereby providing a theoretical justification for preferring it over importance sampling. Our empirical evaluation demonstrates that AND/OR importance sampling is far more accurate than importance sampling in many cases.
Cite
@article{arxiv.1206.3232,
title = {AND/OR Importance Sampling},
author = {Vibhav Gogate and Rina Dechter},
journal= {arXiv preprint arXiv:1206.3232},
year = {2012}
}
Comments
Appears in Proceedings of the Twenty-Fourth Conference on Uncertainty in Artificial Intelligence (UAI2008)