{"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-fast-mixing-models-for-structured","title":"Learning Fast-Mixing Models for Structured Prediction","arxiv_id":"1502.06668","date":"2015-02-24","proceeding":null,"authors":["Jacob Steinhardt","Percy Liang"],"abstract":"Markov Chain Monte Carlo (MCMC) algorithms are often used for approximate\ninference inside learning, but their slow mixing can be difficult to diagnose\nand the approximations can seriously degrade learning. To alleviate these\nissues, we define a new model family using strong Doeblin Markov chains, whose\nmixing times can be precisely controlled by a parameter. We also develop an\nalgorithm to learn such models, which involves maximizing the data likelihood\nunder the induced stationary distribution of these chains. We show empirical\nimprovements on two challenging inference tasks.","url_abs":"http://arxiv.org/abs/1502.06668v1","url_pdf":"http://arxiv.org/pdf/1502.06668v1.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-fast-mixing-models-for-structured","repo_url":"https://worksheets.codalab.org/worksheets/0xc6edf0c9bec643ac9e74418bd6ad4136","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"prediction","task_name":"Prediction"},{"task_slug":"structured-prediction","task_name":"Structured Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}