{"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/from-safe-screening-rules-to-working-sets-for","title":"From safe screening rules to working sets for faster Lasso-type solvers","arxiv_id":"1703.07285","date":"2017-03-21","proceeding":null,"authors":["Mathurin Massias","Alexandre Gramfort","Joseph Salmon"],"abstract":"Convex sparsity-promoting regularizations are ubiquitous in modern\nstatistical learning. By construction, they yield solutions with few non-zero\ncoefficients, which correspond to saturated constraints in the dual\noptimization formulation. Working set (WS) strategies are generic optimization\ntechniques that consist in solving simpler problems that only consider a subset\nof constraints, whose indices form the WS. Working set methods therefore\ninvolve two nested iterations: the outer loop corresponds to the definition of\nthe WS and the inner loop calls a solver for the subproblems. For the Lasso\nestimator a WS is a set of features, while for a Group Lasso it refers to a set\nof groups. In practice, WS are generally small in this context so the\nassociated feature Gram matrix can fit in memory. Here we show that the\nGauss-Southwell rule (a greedy strategy for block coordinate descent\ntechniques) leads to fast solvers in this case. Combined with a working set\nstrategy based on an aggressive use of so-called Gap Safe screening rules, we\npropose a solver achieving state-of-the-art performance on sparse learning\nproblems. Results are presented on Lasso and multi-task Lasso estimators.","url_abs":"http://arxiv.org/abs/1703.07285v2","url_pdf":"http://arxiv.org/pdf/1703.07285v2.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":"from-safe-screening-rules-to-working-sets-for","repo_url":"https://github.com/mathurinm/A5G","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"sparse-learning","task_name":"Sparse Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}