Large-Flip Importance Sampling
Computation
2012-06-26 v1 Artificial Intelligence
Abstract
We propose a new Monte Carlo algorithm for complex discrete distributions. The algorithm is motivated by the N-Fold Way, which is an ingenious event-driven MCMC sampler that avoids rejection moves at any specific state. The N-Fold Way can however get "trapped" in cycles. We surmount this problem by modifying the sampling process. This correction does introduce bias, but the bias is subsequently corrected with a carefully engineered importance sampler.
Cite
@article{arxiv.1206.5239,
title = {Large-Flip Importance Sampling},
author = {Firas Hamze and Nando de Freitas},
journal= {arXiv preprint arXiv:1206.5239},
year = {2012}
}
Comments
Appears in Proceedings of the Twenty-Third Conference on Uncertainty in Artificial Intelligence (UAI2007)