{"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/guaranteed-inference-in-topic-models","title":"Guaranteed inference in topic models","arxiv_id":"1512.03308","date":"2015-12-10","proceeding":null,"authors":["Khoat Than","Tung Doan"],"abstract":"One of the core problems in statistical models is the estimation of a\nposterior distribution. For topic models, the problem of posterior inference\nfor individual texts is particularly important, especially when dealing with\ndata streams, but is often intractable in the worst case. As a consequence,\nexisting methods for posterior inference are approximate and do not have any\nguarantee on neither quality nor convergence rate. In this paper, we introduce\na provably fast algorithm, namely Online Maximum a Posteriori Estimation (OPE),\nfor posterior inference in topic models. OPE has more attractive properties\nthan existing inference approaches, including theoretical guarantees on quality\nand fast rate of convergence to a local maximal/stationary point of the\ninference problem. The discussions about OPE are very general and hence can be\neasily employed in a wide range of contexts. Finally, we employ OPE to design\nthree methods for learning Latent Dirichlet Allocation from text streams or\nlarge corpora. Extensive experiments demonstrate some superior behaviors of OPE\nand of our new learning methods.","url_abs":"http://arxiv.org/abs/1512.03308v2","url_pdf":"http://arxiv.org/pdf/1512.03308v2.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":"guaranteed-inference-in-topic-models","repo_url":"https://github.com/Khoat/OPE","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"topic-models","task_name":"Topic Models"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}