{"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/penalized-estimation-of-directed-acyclic","title":"Penalized Estimation of Directed Acyclic Graphs From Discrete Data","arxiv_id":"1403.2310","date":"2014-03-10","proceeding":null,"authors":["Jiaying Gu","Fei Fu","Qing Zhou"],"abstract":"Bayesian networks, with structure given by a directed acyclic graph (DAG),\nare a popular class of graphical models. However, learning Bayesian networks\nfrom discrete or categorical data is particularly challenging, due to the large\nparameter space and the difficulty in searching for a sparse structure. In this\narticle, we develop a maximum penalized likelihood method to tackle this\nproblem. Instead of the commonly used multinomial distribution, we model the\nconditional distribution of a node given its parents by multi-logit regression,\nin which an edge is parameterized by a set of coefficient vectors with dummy\nvariables encoding the levels of a node. To obtain a sparse DAG, a group norm\npenalty is employed, and a blockwise coordinate descent algorithm is developed\nto maximize the penalized likelihood subject to the acyclicity constraint of a\nDAG. When interventional data are available, our method constructs a causal\nnetwork, in which a directed edge represents a causal relation. We apply our\nmethod to various simulated and real data sets. The results show that our\nmethod is very competitive, compared to many existing methods, in DAG\nestimation from both interventional and high-dimensional observational data.","url_abs":"http://arxiv.org/abs/1403.2310v4","url_pdf":"http://arxiv.org/pdf/1403.2310v4.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":"penalized-estimation-of-directed-acyclic","repo_url":"https://github.com/itsrainingdata/sparsebn","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1403.2310","atlas_url":"https://app.syntology.ai/?focus=1403.2310","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}