{"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/bayesian-coreset-construction-via-greedy","title":"Bayesian Coreset Construction via Greedy Iterative Geodesic Ascent","arxiv_id":"1802.01737","date":"2018-02-05","proceeding":"ICML 2018 7","authors":["Trevor Campbell","Tamara Broderick"],"abstract":"Coherent uncertainty quantification is a key strength of Bayesian methods.\nBut modern algorithms for approximate Bayesian posterior inference often\nsacrifice accurate posterior uncertainty estimation in the pursuit of\nscalability. This work shows that previous Bayesian coreset construction\nalgorithms---which build a small, weighted subset of the data that approximates\nthe full dataset---are no exception. We demonstrate that these algorithms scale\nthe coreset log-likelihood suboptimally, resulting in underestimated posterior\nuncertainty. To address this shortcoming, we develop greedy iterative geodesic\nascent (GIGA), a novel algorithm for Bayesian coreset construction that scales\nthe coreset log-likelihood optimally. GIGA provides geometric decay in\nposterior approximation error as a function of coreset size, and maintains the\nfast running time of its predecessors. The paper concludes with validation of\nGIGA on both synthetic and real datasets, demonstrating that it reduces\nposterior approximation error by orders of magnitude compared with previous\ncoreset constructions.","url_abs":"http://arxiv.org/abs/1802.01737v2","url_pdf":"http://arxiv.org/pdf/1802.01737v2.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":"bayesian-coreset-construction-via-greedy","repo_url":"https://github.com/trevorcampbell/bayesian-coresets","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"uncertainty-quantification","task_name":"Uncertainty Quantification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1802.01737","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}