{"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/robust-probabilistic-modeling-with-bayesian","title":"Robust Probabilistic Modeling with Bayesian Data Reweighting","arxiv_id":"1606.03860","date":"2016-06-13","proceeding":"ICML 2017 8","authors":["Yixin Wang","Alp Kucukelbir","David M. Blei"],"abstract":"Probabilistic models analyze data by relying on a set of assumptions. Data\nthat exhibit deviations from these assumptions can undermine inference and\nprediction quality. Robust models offer protection against mismatch between a\nmodel's assumptions and reality. We propose a way to systematically detect and\nmitigate mismatch of a large class of probabilistic models. The idea is to\nraise the likelihood of each observation to a weight and then to infer both the\nlatent variables and the weights from data. Inferring the weights allows a\nmodel to identify observations that match its assumptions and down-weight\nothers. This enables robust inference and improves predictive accuracy. We\nstudy four different forms of mismatch with reality, ranging from missing\nlatent groups to structure misspecification. A Poisson factorization analysis\nof the Movielens 1M dataset shows the benefits of this approach in a practical\nscenario.","url_abs":"http://arxiv.org/abs/1606.03860v3","url_pdf":"http://arxiv.org/pdf/1606.03860v3.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":"robust-probabilistic-modeling-with-bayesian","repo_url":"https://github.com/yixinwang/robust-rpm-public","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1606.03860","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}