Abstrakti

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.
AlkuperäiskieliEnglanti
OtsikkoProceedings of The 25th International Conference on Artificial Intelligence and Statistics
KustantajaJMLR
Sivut1786-1804
TilaJulkaistu - 2022
OKM-julkaisutyyppiA4 Artikkeli konferenssijulkaisussa
TapahtumaInternational Conference on Artificial Intelligence and Statistics - Valencia, Espanja
Kesto: 28 maalisk. 202230 maalisk. 2022
Konferenssinumero: 25

Julkaisusarja

NimiProceedings of Machine Learning Research
KustantajaPMLR
Vuosikerta151
ISSN (elektroninen)2640-3498

Conference

ConferenceInternational Conference on Artificial Intelligence and Statistics
LyhennettäAISTATS
Maa/AlueEspanja
KaupunkiValencia
Ajanjakso28/03/202230/03/2022

Sormenjälki

Sukella tutkimusaiheisiin 'Parallel MCMC Without Embarrassing Failures'. Ne muodostavat yhdessä ainutlaatuisen sormenjäljen.

Siteeraa tätä