![](https://d3ilqtpdwi981i.cloudfront.net/FNDQDAhtWprcfI5neCOPlTa46LI=/425x550/smart/https://bepress-attached-resources.s3.amazonaws.com/uploads/00/f3/c1/00f3c111-b619-4ea8-9bca-d4655f07ccff/thumbnail_BPFile%20object.jpg)
Unpublished Paper
Monte Carlo MCMC: Efficient Inference by Sampling Factors
(2012)
Abstract
Discriminative graphical models such as conditional random fields and Markov logic net- works have achieved state of the art results in a variety of NLP and IE tasks including coreference and relation extraction. Increasingly, automated knowledge extraction is demanding models with more complex structure— higher tree-width, larger fan-out, more features, more data—rendering even approximate inference methods such as MCMC inefficient. In this paper we propose a new MCMC sampling scheme where transition probabilities are approximated. We demonstrate that our method converges more quickly than a traditional MCMC sampler for both marginal and MAP inference. For a task of author coreference over 5 million mentions, we achieve a speedup of 13 over regular MCMC inference.
Disciplines
Publication Date
2012
Comments
This is the pre-published version harvested from CIIR.
Citation Information
Sameer Singh, Michael Wick and Andrew McCallum. "Monte Carlo MCMC: Efficient Inference by Sampling Factors" (2012) Available at: http://works.bepress.com/andrew_mccallum/59/