Large-Flip Importance Sampling

UAI 2007 pp. 167-174

doi:10.5555/3020488.3020509 /uai/2007/hamze2007uai-large/

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.

UAI Semantic Scholar

Cite

Text

Hamze and de Freitas. "Large-Flip Importance Sampling." Conference on Uncertainty in Artificial Intelligence, 2007. doi:10.5555/3020488.3020509

Markdown

[Hamze and de Freitas. "Large-Flip Importance Sampling." Conference on Uncertainty in Artificial Intelligence, 2007.](https://mlanthology.org/uai/2007/hamze2007uai-large/) doi:10.5555/3020488.3020509

BibTeX

@inproceedings{hamze2007uai-large,
  title     = {{Large-Flip Importance Sampling}},
  author    = {Hamze, Firas and de Freitas, Nando},
  booktitle = {Conference on Uncertainty in Artificial Intelligence},
  year      = {2007},
  pages     = {167-174},
  doi       = {10.5555/3020488.3020509},
  url       = {https://mlanthology.org/uai/2007/hamze2007uai-large/}
}