close this message
arXiv smileybones

arXiv Is Hiring a DevOps Engineer

Work on one of the world's most important websites and make an impact on open science.

View Jobs
Skip to main content
Cornell University

arXiv Is Hiring a DevOps Engineer

View Jobs
We gratefully acknowledge support from the Simons Foundation, member institutions, and all contributors. Donate
arxiv logo > stat > arXiv:2202.11154

Help | Advanced Search

arXiv logo
Cornell University Logo

quick links

  • Login
  • Help Pages
  • About

Statistics > Machine Learning

arXiv:2202.11154 (stat)
[Submitted on 22 Feb 2022 (v1), last revised 29 Mar 2022 (this version, v2)]

Title:Parallel MCMC Without Embarrassing Failures

Authors:Daniel Augusto de Souza, Diego Mesquita, Samuel Kaski, Luigi Acerbi
View a PDF of the paper titled Parallel MCMC Without Embarrassing Failures, by Daniel Augusto de Souza and 3 other authors
View PDF
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.
Comments: To appear in the 25th International Conference on Artificial Intelligence and Statistics (AISTATS 2022). For associated code, see this https URL
Subjects: Machine Learning (stat.ML); Machine Learning (cs.LG); Methodology (stat.ME)
Cite as: arXiv:2202.11154 [stat.ML]
  (or arXiv:2202.11154v2 [stat.ML] for this version)
  https://doi.org/10.48550/arXiv.2202.11154
arXiv-issued DOI via DataCite

Submission history

From: Daniel Augusto de Souza [view email]
[v1] Tue, 22 Feb 2022 20:17:46 UTC (1,823 KB)
[v2] Tue, 29 Mar 2022 23:50:34 UTC (1,823 KB)
Full-text links:

Access Paper:

    View a PDF of the paper titled Parallel MCMC Without Embarrassing Failures, by Daniel Augusto de Souza and 3 other authors
  • View PDF
  • TeX Source
  • Other Formats
license icon view license
Current browse context:
stat.ML
< prev   |   next >
new | recent | 2022-02
Change to browse by:
cs
cs.LG
stat
stat.ME

References & Citations

  • NASA ADS
  • Google Scholar
  • Semantic Scholar
a export BibTeX citation Loading...

BibTeX formatted citation

×
Data provided by:

Bookmark

BibSonomy logo Reddit logo

Bibliographic and Citation Tools

Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)

Code, Data and Media Associated with this Article

alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)

Demos

Replicate (What is Replicate?)
Hugging Face Spaces (What is Spaces?)
TXYZ.AI (What is TXYZ.AI?)

Recommenders and Search Tools

Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
  • Author
  • Venue
  • Institution
  • Topic

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)
  • About
  • Help
  • contact arXivClick here to contact arXiv Contact
  • subscribe to arXiv mailingsClick here to subscribe Subscribe
  • Copyright
  • Privacy Policy
  • Web Accessibility Assistance
  • arXiv Operational Status
    Get status notifications via email or slack