Likelihood inflating sampling algorithm

Markov Chain Monte Carlo (MCMC) sampling from a posterior distribution corresponding to a massive data set can be computationally prohibitive as producing one sample requires a number of operations that is linear in the data size. In this article we introduce a new communication‐free parallel method, the “Likelihood Inflating Sampling Algorithm (LISA),” that significantly reduces computational costs by randomly splitting the data set into smaller subsets and running MCMC methods “independently” in parallel on each subset using different processors. Each processor will be used to run an MCMC chain that samples sub‐posterior distributions which are defined using an “inflated” likelihood function. We develop a strategy for combining the draws from different sub‐posteriors to study the full posterior of the Bayesian Additive Regression Trees (BART) model. The performance of the method is tested using simulated data and a large socio‐economic study. The Canadian Journal of Statistics 46: 147–175; 2018 © 2017 Statistical Society of Canada

• Publication year

  2016

• Venue

  Canadian Journal of Statistics

• Publication date

  2016-05-06

• Fields of study

  Mathematics, Computer Science

• Identifiers
  DOI 10.1002/cjs.11343arXiv 1605.02113
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Bayesian Computation Via Markov Chain Monte Carlo

  2014cited by this paper
• Asymptotically Exact, Embarrassingly Parallel MCMC

  2013cited by this paper
• Efficient Metropolis–Hastings Proposal Mechanisms for Bayesian Regression Tree Models

  2013cited by this paper
• Parallelizing MCMC via Weierstrass Sampler

  2013cited by this paper
• bartMachine: Machine Learning with Bayesian Additive Regression Trees

  2013cited by this paper
• Handbook of Markov Chain Monte Carlo

  2011cited by this paper
• SciNet: Lessons Learned from Building a Power-efficient Top-20 System and Data Centre

  2010cited by this paper
• BART: Bayesian Additive Regression Trees

  2008cited by this paper
• Bayesian Additive Regression Trees

  2006cited by this paper
• Parallel Bayesian Computation

  2005cited by this paper
• Population Markov Chain Monte Carlo

  2004cited by this paper
• Parallel computing and Monte Carlo algorithms

  1999cited by this paper
• Bayesian CART Model Search

  1998cited by this paper
• Journal of the American Statistical Association Bayesian Inference for Logistic Models Using Pólya–gamma Latent Variables Bayesian Inference for Logistic Models Using P ´ Olya–gamma Latent Variables

  year unknowncited by this paper

• Composite method for fast computation of individual level spatial epidemic models

  2025cites this paper
• Type I Tobit Bayesian Additive Regression Trees for censored outcome regression

  2022cites this paper
• Approximate Methods for Bayesian Computation

  2022cites this paper
• Reflections on Bayesian inference and Markov chain Monte Carlo

  2022cites this paper
• SwISS: A scalable Markov chain Monte Carlo divide‐and‐conquer strategy

  2022cites this paper
• MCMC-driven importance samplers

  2021cites this paper
• Distributed Bayesian Varying Coefficient Modeling Using a Gaussian Process Prior

  2020cites this paper
• Sparse Variational Inference: Bayesian Coresets from Scratch

  2019cites this paper
• Bayesian additive regression trees and the General BART model

  2019cites this paper
• Finding our Way in the Dark: Approximate MCMC for Approximate Bayesian Methods

  2019cites this paper
• Consensus Monte Carlo for Random Subsets Using Shared Anchors

  2019cites this paper
• State-of-the-BART: Simple Bayesian Tree Algorithms for Prediction and Causal Inference

  2019cites this paper
• The BART R package

  2019cites this paper
• Bayesian Additive Regression Trees ( BART )

  2019cites this paper
• Method G: Uncertainty Quantification for Distributed Data Problems Using Generalized Fiducial Inference

  2018cites this paper
• PASS-GLM: polynomial approximate sufficient statistics for scalable Bayesian GLM inference

  2017influential citation
• Comment by Oksana A. Chkrebtii 1 Population Mcmc with Efficient Tree Proposals 2 Prior for Model Selection Comment

  2017influential citation
• Automated Scalable Bayesian Inference via Hilbert Coresets

  2017cites this paper
• Adaptive Component-Wise Multiple-Try Metropolis Sampling

  2016cites this paper
• Comment on Article by Matthew T . Pratola

  2016cites this paper
• Contributed discussion on article by Pratola [Comment on "M.T. Pratola, Efficient metropolis-hastings proposal mechanisms for Bayesian regression tree models"]

  2016influential citation
• Coresets for Scalable Bayesian Logistic Regression

  2016influential citation
• Contributed Discussion on Article by Pratola

  2016influential citation