{"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/partition-mcmc-for-inference-on-acyclic","title":"Partition MCMC for inference on acyclic digraphs","arxiv_id":"1504.05006","date":"2015-04-20","proceeding":null,"authors":["Jack Kuipers","Giusi Moffa"],"abstract":"Acyclic digraphs are the underlying representation of Bayesian networks, a\nwidely used class of probabilistic graphical models. Learning the underlying\ngraph from data is a way of gaining insights about the structural properties of\na domain. Structure learning forms one of the inference challenges of\nstatistical graphical models.\n  MCMC methods, notably structure MCMC, to sample graphs from the posterior\ndistribution given the data are probably the only viable option for Bayesian\nmodel averaging. Score modularity and restrictions on the number of parents of\neach node allow the graphs to be grouped into larger collections, which can be\nscored as a whole to improve the chain's convergence. Current examples of\nalgorithms taking advantage of grouping are the biased order MCMC, which acts\non the alternative space of permuted triangular matrices, and non ergodic edge\nreversal moves.\n  Here we propose a novel algorithm, which employs the underlying combinatorial\nstructure of DAGs to define a new grouping. As a result convergence is improved\ncompared to structure MCMC, while still retaining the property of producing an\nunbiased sample. Finally the method can be combined with edge reversal moves to\nimprove the sampler further.","url_abs":"http://arxiv.org/abs/1504.05006v2","url_pdf":"http://arxiv.org/pdf/1504.05006v2.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":"partition-mcmc-for-inference-on-acyclic","repo_url":"https://github.com/annlia/partitionMCMC","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1504.05006","atlas_url":"https://app.syntology.ai/?focus=1504.05006","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}