Particle MCMC algorithms and architectures for accelerating inference in state-space models
File(s)1-s2.0-S0888613X16302092-main.pdf (1021.59 KB)
Published version
Author(s)
Bouganis, C
Mingas, G
Bottolo, L
Type
Journal Article
Abstract
Particle Markov Chain Monte Carlo (pMCMC) is a stochastic algorithm designed to generate samples from a prob-
ability distribution, when the density of the distribution does not admit a closed form expression. pMCMC is most
commonly used to sample from the Bayesian posterior distribution in State-Space Models (SSMs), a class of prob-
abilistic models used in numerous scientific applications. Nevertheless, this task is prohibitive when dealing with
complex SSMs with massive data, due to the high computational cost of pMCMC and its poor performance when the
posterior exhibits multi-modality. This paper aims to address both issues by: 1) Proposing a novel pMCMC algorithm
(denoted ppMCMC), which uses multiple Markov chains (instead of the one used by pMCMC) to improve sampling
ability distribution, when the density of the distribution does not admit a closed form expression. pMCMC is most
commonly used to sample from the Bayesian posterior distribution in State-Space Models (SSMs), a class of prob-
abilistic models used in numerous scientific applications. Nevertheless, this task is prohibitive when dealing with
complex SSMs with massive data, due to the high computational cost of pMCMC and its poor performance when the
posterior exhibits multi-modality. This paper aims to address both issues by: 1) Proposing a novel pMCMC algorithm
(denoted ppMCMC), which uses multiple Markov chains (instead of the one used by pMCMC) to improve sampling
Date Issued
2016-11-14
Date Acceptance
2016-10-25
Citation
International Journal of Approximate Reasoning, 2016, 83, pp.413-433
ISSN
1873-4731
Publisher
Elsevier
Start Page
413
End Page
433
Journal / Book Title
International Journal of Approximate Reasoning
Volume
83
Copyright Statement
© 2016 The Authors. Published by Elsevier Inc. This is an open access article under the CC
BY license (http://creativecommons.org/licenses/by/4.0/).
BY license (http://creativecommons.org/licenses/by/4.0/).
Sponsor
Engineering & Physical Science Research Council (EPSRC)
Wellcome Trust
Grant Number
EP/I012036/1
097816/Z/11/A
Subjects
Science & Technology
Technology
Computer Science, Artificial Intelligence
Computer Science
Markov Chain Monte Carlo
Particle filter
Field programmable gate array
Bayesian inference
Hardware acceleration
CHAIN MONTE-CARLO
MARKOV
PERSPECTIVES
Artificial Intelligence & Image Processing
0103 Numerical And Computational Mathematics
0104 Statistics
0801 Artificial Intelligence And Image Processing
Publication Status
Published