{"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/constant-time-predictive-distributions-for","title":"Constant-Time Predictive Distributions for Gaussian Processes","arxiv_id":"1803.06058","date":"2018-03-16","proceeding":"ICML 2018 7","authors":["Geoff Pleiss","Jacob R. Gardner","Kilian Q. Weinberger","Andrew Gordon Wilson"],"abstract":"One of the most compelling features of Gaussian process (GP) regression is\nits ability to provide well-calibrated posterior distributions. Recent advances\nin inducing point methods have sped up GP marginal likelihood and posterior\nmean computations, leaving posterior covariance estimation and sampling as the\nremaining computational bottlenecks. In this paper we address these\nshortcomings by using the Lanczos algorithm to rapidly approximate the\npredictive covariance matrix. Our approach, which we refer to as LOVE (LanczOs\nVariance Estimates), substantially improves time and space complexity. In our\nexperiments, LOVE computes covariances up to 2,000 times faster and draws\nsamples 18,000 times faster than existing methods, all without sacrificing\naccuracy.","url_abs":"http://arxiv.org/abs/1803.06058v4","url_pdf":"http://arxiv.org/pdf/1803.06058v4.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":"constant-time-predictive-distributions-for","repo_url":"https://github.com/cornellius-gp/gpytorch/blob/master/examples/02_Scalable_Exact_GPs/Simple_GP_Regression_With_LOVE_Fast_Variances_and_Sampling.ipynb","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"gaussian-processes","task_name":"Gaussian Processes"},{"task_slug":"regression-1","task_name":"regression"}],"methods":[{"method_slug":"gaussian-process","method_name":"Gaussian Process"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1803.06058","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}