{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/divide-and-conquer-with-sequential-monte","title":"Divide-and-Conquer with Sequential Monte Carlo","arxiv_id":"1406.4993","date":"2014-06-19","proceeding":null,"authors":["Fredrik Lindsten","Adam M. Johansen","Christian A. Naesseth","Bonnie Kirkpatrick","Thomas B. Schön","John Aston","Alexandre Bouchard-Côté"],"abstract":"We propose a novel class of Sequential Monte Carlo (SMC) algorithms,\nappropriate for inference in probabilistic graphical models. This class of\nalgorithms adopts a divide-and-conquer approach based upon an auxiliary\ntree-structured decomposition of the model of interest, turning the overall\ninferential task into a collection of recursively solved sub-problems. The\nproposed method is applicable to a broad class of probabilistic graphical\nmodels, including models with loops. Unlike a standard SMC sampler, the\nproposed Divide-and-Conquer SMC employs multiple independent populations of\nweighted particles, which are resampled, merged, and propagated as the method\nprogresses. We illustrate empirically that this approach can outperform\nstandard methods in terms of the accuracy of the posterior expectation and\nmarginal likelihood approximations. Divide-and-Conquer SMC also opens up novel\nparallel implementation options and the possibility of concentrating the\ncomputational effort on the most challenging sub-problems. We demonstrate its\nperformance on a Markov random field and on a hierarchical logistic regression\nproblem.","url_abs":"http://arxiv.org/abs/1406.4993v2","url_pdf":"http://arxiv.org/pdf/1406.4993v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"divide-and-conquer-with-sequential-monte","repo_url":"https://github.com/alexandrebouchard/multilevelSMC","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"divide-and-conquer-with-sequential-monte","repo_url":"https://github.com/alexandrebouchard/multilevelSMC-experiments","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"divide-and-conquer-with-sequential-monte","repo_url":"https://github.com/tbrx/compiled-inference","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1406.4993","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}