{"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/exploiting-structure-for-fast-kernel-learning","title":"Exploiting Structure for Fast Kernel Learning","arxiv_id":"1808.03351","date":"2018-08-09","proceeding":null,"authors":["Trefor W. Evans","Prasanth B. Nair"],"abstract":"We propose two methods for exact Gaussian process (GP) inference and learning\non massive image, video, spatial-temporal, or multi-output datasets with\nmissing values (or \"gaps\") in the observed responses. The first method ignores\nthe gaps using sparse selection matrices and a highly effective low-rank\npreconditioner is introduced to accelerate computations. The second method\nintroduces a novel approach to GP training whereby response values are inferred\non the gaps before explicitly training the model. We find this second approach\nto be greatly advantageous for the class of problems considered. Both of these\nnovel approaches make extensive use of Kronecker matrix algebra to design\nmassively scalable algorithms which have low memory requirements. We\ndemonstrate exact GP inference for a spatial-temporal climate modelling problem\nwith 3.7 million training points as well as a video reconstruction problem with\n1 billion points.","url_abs":"http://arxiv.org/abs/1808.03351v1","url_pdf":"http://arxiv.org/pdf/1808.03351v1.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":"exploiting-structure-for-fast-kernel-learning","repo_url":"https://github.com/treforevans/gp_grid","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"missing-values","task_name":"Missing Values"},{"task_slug":"video-reconstruction","task_name":"Video Reconstruction"}],"methods":[{"method_slug":"gaussian-process","method_name":"Gaussian Process"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}