{"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/faster-greedy-map-inference-for-determinantal","title":"Faster Greedy MAP Inference for Determinantal Point Processes","arxiv_id":"1703.03389","date":"2017-03-09","proceeding":"ICML 2017 8","authors":["Insu Han","Prabhanjan Kambadur","KyoungSoo Park","Jinwoo Shin"],"abstract":"Determinantal point processes (DPPs) are popular probabilistic models that\narise in many machine learning tasks, where distributions of diverse sets are\ncharacterized by matrix determinants. In this paper, we develop fast algorithms\nto find the most likely configuration (MAP) of large-scale DPPs, which is\nNP-hard in general. Due to the submodular nature of the MAP objective, greedy\nalgorithms have been used with empirical success. Greedy implementations\nrequire computation of log-determinants, matrix inverses or solving linear\nsystems at each iteration. We present faster implementations of the greedy\nalgorithms by utilizing the complementary benefits of two log-determinant\napproximation schemes: (a) first-order expansions to the matrix log-determinant\nfunction and (b) high-order expansions to the scalar log function with\nstochastic trace estimators. In our experiments, our algorithms are orders of\nmagnitude faster than their competitors, while sacrificing marginal accuracy.","url_abs":"http://arxiv.org/abs/1703.03389v2","url_pdf":"http://arxiv.org/pdf/1703.03389v2.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":"faster-greedy-map-inference-for-determinantal","repo_url":"https://github.com/insuhan/fastdppmap","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"point-processes","task_name":"Point Processes"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1703.03389","atlas_url":"https://app.syntology.ai/?focus=1703.03389","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}