Accelerated Adaptive Markov Chain for Partition Function Computation
Abstract
We propose a novel Adaptive Markov Chain Monte Carlo algorithm to compute the partition function. In particular, we show how to accelerate a flat histogram sampling technique by significantly reducing the number of ``null moves'' in the chain, while maintaining asymptotic convergence properties. Our experiments show that our method converges quickly to highly accurate solutions on a range of benchmark instances, outperforming other state-of-the-art methods such as IJGP, TRW, and Gibbs sampling both in run-time and accuracy. We also show how obtaining a so-called density of states distribution allows for efficient weight learning in Markov Logic theories.
Cite
Text
Ermon et al. "Accelerated Adaptive Markov Chain for Partition Function Computation." Neural Information Processing Systems, 2011.Markdown
[Ermon et al. "Accelerated Adaptive Markov Chain for Partition Function Computation." Neural Information Processing Systems, 2011.](https://mlanthology.org/neurips/2011/ermon2011neurips-accelerated/)BibTeX
@inproceedings{ermon2011neurips-accelerated,
title = {{Accelerated Adaptive Markov Chain for Partition Function Computation}},
author = {Ermon, Stefano and Gomes, Carla P. and Sabharwal, Ashish and Selman, Bart},
booktitle = {Neural Information Processing Systems},
year = {2011},
pages = {2744-2752},
url = {https://mlanthology.org/neurips/2011/ermon2011neurips-accelerated/}
}