{"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/variable-selection-for-gaussian-processes-via","title":"Variable selection for Gaussian processes via sensitivity analysis of the posterior predictive distribution","arxiv_id":"1712.08048","date":"2017-12-21","proceeding":null,"authors":["Topi Paananen","Juho Piironen","Michael Riis Andersen","Aki Vehtari"],"abstract":"Variable selection for Gaussian process models is often done using automatic\nrelevance determination, which uses the inverse length-scale parameter of each\ninput variable as a proxy for variable relevance. This implicitly determined\nrelevance has several drawbacks that prevent the selection of optimal input\nvariables in terms of predictive performance. To improve on this, we propose\ntwo novel variable selection methods for Gaussian process models that utilize\nthe predictions of a full model in the vicinity of the training points and\nthereby rank the variables based on their predictive relevance. Our empirical\nresults on synthetic and real world data sets demonstrate improved variable\nselection compared to automatic relevance determination in terms of variability\nand predictive performance.","url_abs":"http://arxiv.org/abs/1712.08048v3","url_pdf":"http://arxiv.org/pdf/1712.08048v3.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":"variable-selection-for-gaussian-processes-via","repo_url":"https://github.com/topipa/GP_varsel_KL_VAR","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null},{"paper_slug":"variable-selection-for-gaussian-processes-via","repo_url":"https://github.com/topipa/gp-varsel-kl-var","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"gaussian-processes","task_name":"Gaussian Processes"},{"task_slug":"sensitivity","task_name":"Sensitivity"},{"task_slug":"variable-selection","task_name":"Variable Selection"}],"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}