Parallel Markov chain Monte Carlo simulation by pre-fetching
Parallel Markov chain Monte Carlo simulation by pre-fetching
复制标题
DOI：
10.1198/106186006x100579
复制
发表时间：
2006-03-01
影响因子：
2.4
通讯作者：
Brockwell, AE
中科院分区：
文献类型：
--
作者：
Brockwell, AE
近年来，并行处理已经成为研究人员广泛使用的工具。它可以以一种明显的方式应用于蒙特卡罗模拟的环境中，但除了并行生成多个链的自然方法外，对马尔可夫链蒙特卡罗(MCMC)算法进行“并行化”的技术并不那么明显。虽然并行链的生成通常是最简单的方法，但在老化是一个严重问题的情况下，通常希望使用并行化来加速单链的生成。本文简要讨论了现有的一些MCMC算法的并行化方法，并提出了一种新的“预取”算法来实现单链生成的并行化。
In recent years, parallel processing has become widely available to researchers. It can be applied in an obvious way in the context of Monte Carlo simulation, but techniques for "parallelizing" Markov chain Monte Carlo (MCMC) algorithms are not so obvious, apart from the natural approach of generating multiple chains in parallel. Although generation of parallel chains is generally the easiest approach, in cases where burn-in is a serious problem, it is often desirable to use parallelization to speed up generation of a single chain. This article briefly discusses some existing methods for parallelization of MCMC algorithms, and proposes a new "pre-fetching" algorithm to parallelize generation of a single chain.