{"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/structured-nonlinear-variable-selection","title":"Structured nonlinear variable selection","arxiv_id":"1805.06258","date":"2018-05-16","proceeding":null,"authors":["Magda Gregorová","Alexandros Kalousis","Stéphane Marchand-Maillet"],"abstract":"We investigate structured sparsity methods for variable selection in\nregression problems where the target depends nonlinearly on the inputs. We\nfocus on general nonlinear functions not limiting a priori the function space\nto additive models. We propose two new regularizers based on partial\nderivatives as nonlinear equivalents of group lasso and elastic net. We\nformulate the problem within the framework of learning in reproducing kernel\nHilbert spaces and show how the variational problem can be reformulated into a\nmore practical finite dimensional equivalent. We develop a new algorithm\nderived from the ADMM principles that relies solely on closed forms of the\nproximal operators. We explore the empirical properties of our new algorithm\nfor Nonlinear Variable Selection based on Derivatives (NVSD) on a set of\nexperiments and confirm favourable properties of our structured-sparsity models\nand the algorithm in terms of both prediction and variable selection accuracy.","url_abs":"http://arxiv.org/abs/1805.06258v1","url_pdf":"http://arxiv.org/pdf/1805.06258v1.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":"structured-nonlinear-variable-selection","repo_url":"https://bitbucket.org/dmmlgeneva/nvsd_uai2018","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"additive-models","task_name":"Additive models"},{"task_slug":"variable-selection","task_name":"Variable Selection"}],"methods":[{"method_slug":"admm","method_name":"ADMM"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}