Accelerating MCMC via Parallel Predictive Prefetching
Abstract
Parallel predictive prefetching is a new framework for accelerating a large class of widely-used Markov chain Monte Carlo (MCMC) algorithms. It speculatively evaluates many potential steps of an MCMC chain in parallel while exploiting fast, iterative approximations to the target density. This can accelerate sampling from target distributions in Bayesian inference problems. Our approach takes advantage of whatever parallel resources are available, but produces results exactly equivalent to standard serial execution. In the initial burn-in phase of chain evaluation, we achieve speedup close to linear in the number of available cores.
Cite
Text
Angelino et al. "Accelerating MCMC via Parallel Predictive Prefetching." Conference on Uncertainty in Artificial Intelligence, 2014.Markdown
[Angelino et al. "Accelerating MCMC via Parallel Predictive Prefetching." Conference on Uncertainty in Artificial Intelligence, 2014.](https://mlanthology.org/uai/2014/angelino2014uai-accelerating/)BibTeX
@inproceedings{angelino2014uai-accelerating,
title = {{Accelerating MCMC via Parallel Predictive Prefetching}},
author = {Angelino, Elaine and Kohler, Eddie and Waterland, Amos and Seltzer, Margo I. and Adams, Ryan P.},
booktitle = {Conference on Uncertainty in Artificial Intelligence},
year = {2014},
pages = {22-31},
url = {https://mlanthology.org/uai/2014/angelino2014uai-accelerating/}
}