Parallel MCMC Without Embarrassing Failures

Loading...
Thumbnail Image

Access rights

openAccess

URL

Journal Title

Journal ISSN

Volume Title

A4 Artikkeli konferenssijulkaisussa

Date

2022

Major/Subject

Mcode

Degree programme

Language

en

Pages

1786-1804

Series

Proceedings of The 25th International Conference on Artificial Intelligence and Statistics, Proceedings of Machine Learning Research, Volume 151

Abstract

Embarrassingly parallel Markov Chain Monte Carlo (MCMC) exploits parallel computing to scale Bayesian inference to large datasets by using a two-step approach. First, MCMC is run in parallel on (sub)posteriors defined on data partitions. Then, a server combines local results. While efficient, this framework is very sensitive to the quality of subposterior sampling. Common sampling problems such as missing modes or misrepresentation of low-density regions are amplified – instead of being corrected – in the combination phase, leading to catastrophic failures. In this work, we propose a novel combination strategy to mitigate this issue. Our strategy, Parallel Active Inference (PAI), leverages Gaussian Process (GP) surrogate modeling and active learning. After fitting GPs to subposteriors, PAI (i) shares information between GP surrogates to cover missing modes; and (ii) uses active sampling to individually refine subposterior approximations. We validate PAI in challenging benchmarks, including heavy-tailed and multi-modal posteriors and a real-world application to computational neuroscience. Empirical results show that PAI succeeds where previous methods catastrophically fail, with a small communication overhead.

Description

Keywords

Other note

Citation

Augusto de Souza, D, Parente Paiva Mesquita, D, Kaski, S & Acerbi, L 2022, Parallel MCMC Without Embarrassing Failures . in Proceedings of The 25th International Conference on Artificial Intelligence and Statistics . Proceedings of Machine Learning Research, vol. 151, JMLR, pp. 1786-1804, International Conference on Artificial Intelligence and Statistics, Valencia, Spain, 28/03/2022 . < https://proceedings.mlr.press/v151/de-souza22a/de-souza22a.pdf >