{"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/bayesian-structure-learning-by-recursive","title":"Bayesian Structure Learning by Recursive Bootstrap","arxiv_id":"1809.04828","date":"2018-09-13","proceeding":"NeurIPS 2018 12","authors":["Raanan Y. Rohekar","Yaniv Gurwicz","Shami Nisimov","Guy Koren","Gal Novik"],"abstract":"We address the problem of Bayesian structure learning for domains with\nhundreds of variables by employing non-parametric bootstrap, recursively. We\npropose a method that covers both model averaging and model selection in the\nsame framework. The proposed method deals with the main weakness of\nconstraint-based learning---sensitivity to errors in the independence\ntests---by a novel way of combining bootstrap with constraint-based learning.\nEssentially, we provide an algorithm for learning a tree, in which each node\nrepresents a scored CPDAG for a subset of variables and the level of the node\ncorresponds to the maximal order of conditional independencies that are encoded\nin the graph. As higher order independencies are tested in deeper recursive\ncalls, they benefit from more bootstrap samples, and therefore more resistant\nto the curse-of-dimensionality. Moreover, the re-use of stable low order\nindependencies allows greater computational efficiency. We also provide an\nalgorithm for sampling CPDAGs efficiently from their posterior given the\nlearned tree. We empirically demonstrate that the proposed algorithm scales\nwell to hundreds of variables, and learns better MAP models and more reliable\ncausal relationships between variables, than other state-of-the-art-methods.","url_abs":"http://arxiv.org/abs/1809.04828v1","url_pdf":"http://arxiv.org/pdf/1809.04828v1.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":"bayesian-structure-learning-by-recursive","repo_url":"https://github.com/IntelLabs/causality-lab","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"computational-efficiency","task_name":"Computational Efficiency"},{"task_slug":"model-selection","task_name":"Model Selection"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1809.04828","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}