Journal of Computational and Graphical Statistics · 2016 · 33 citations · 49 references
Natural SciencesBiostatisticsStatistical ModelingStatisticsStochastic ModelingBayesian ComputationBayesian InferenceBayesian StatisticsInferential StatisticsMonte Carlo MethodsStatistical InferenceConditional ExpectationGraphical ModelsHierarchical Logistic RegressionLogistic RegressionMarginal LikelihoodMarginal Likelihood MaximizationMarkov Random Field ModelsMarkov Random Fields
We propose a novel class of Sequential Monte Carlo (SMC) algorithms, appropriate for inference in probabilistic graphical models. This class of algorithms adopts a divide-and-conquer approach based upon an auxiliary tree-structured decomposition of the model of interest, turning the overall inferential task into a collection of recursively solved subproblems. The proposed method is applicable to a broad class of probabilistic graphical models, including models with loops. Unlike a standard SMC sampler, the proposed divide-and-conquer SMC employs multiple independent populations of weighted particles, which are resampled, merged, and propagated as the method progresses. We illustrate empirically that this approach can outperform standard methods in terms of the accuracy of the posterior expectation and marginal likelihood approximations. Divide-and-conquer SMC also opens up novel parallel implementation options and the possibility of concentrating the computational effort on the most challenging subproblems. We demonstrate its performance on a Markov random field and on a hierarchical logistic regression problem. Supplementary materials including proofs and additional numerical results are available online.
49
Proceedings of the 24th international conference on Machine learning
2007 · 11.7K citations
Pattern Recognition and Machine Learning
Kybernetes · 2007 · 8.4K citations
<i>Stan</i> : A Probabilistic Programming Language
Bob Carpenter, Andrew Gelman, Matthew D. Hoffman et al. · Journal of Statistical Software · 2017 · 7.4K citations · Full text