{"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/learning-certifiably-optimal-rule-lists-for","title":"Learning Certifiably Optimal Rule Lists for Categorical Data","arxiv_id":"1704.01701","date":"2017-04-06","proceeding":null,"authors":["Elaine Angelino","Nicholas Larus-Stone","Daniel Alabi","Margo Seltzer","Cynthia Rudin"],"abstract":"We present the design and implementation of a custom discrete optimization\ntechnique for building rule lists over a categorical feature space. Our\nalgorithm produces rule lists with optimal training performance, according to\nthe regularized empirical risk, with a certificate of optimality. By leveraging\nalgorithmic bounds, efficient data structures, and computational reuse, we\nachieve several orders of magnitude speedup in time and a massive reduction of\nmemory consumption. We demonstrate that our approach produces optimal rule\nlists on practical problems in seconds. Our results indicate that it is\npossible to construct optimal sparse rule lists that are approximately as\naccurate as the COMPAS proprietary risk prediction tool on data from Broward\nCounty, Florida, but that are completely interpretable. This framework is a\nnovel alternative to CART and other decision tree methods for interpretable\nmodeling.","url_abs":"http://arxiv.org/abs/1704.01701v4","url_pdf":"http://arxiv.org/pdf/1704.01701v4.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":"learning-certifiably-optimal-rule-lists-for","repo_url":"https://github.com/nlarusstone/corels","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"learning-certifiably-optimal-rule-lists-for","repo_url":"https://github.com/corels/rcppcorels","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"learning-certifiably-optimal-rule-lists-for","repo_url":"https://github.com/eddelbuettel/rcppcorels","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"learning-certifiably-optimal-rule-lists-for","repo_url":"https://github.com/fingoldin/pycorels","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"learning-certifiably-optimal-rule-lists-for","repo_url":"https://github.com/saligrama/rcorels","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":{"status":"unanswered"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1704.01701","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}