{"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/online-variance-reduction-for-stochastic","title":"Online Variance Reduction for Stochastic Optimization","arxiv_id":"1802.04715","date":"2018-02-13","proceeding":null,"authors":["Zalán Borsos","Andreas Krause","Kfir. Y. Levy"],"abstract":"Modern stochastic optimization methods often rely on uniform sampling which\nis agnostic to the underlying characteristics of the data. This might degrade\nthe convergence by yielding estimates that suffer from a high variance. A\npossible remedy is to employ non-uniform importance sampling techniques, which\ntake the structure of the dataset into account. In this work, we investigate a\nrecently proposed setting which poses variance reduction as an online\noptimization problem with bandit feedback. We devise a novel and efficient\nalgorithm for this setting that finds a sequence of importance sampling\ndistributions competitive with the best fixed distribution in hindsight, the\nfirst result of this kind. While we present our method for sampling datapoints,\nit naturally extends to selecting coordinates or even blocks of thereof.\nEmpirical validations underline the benefits of our method in several settings.","url_abs":"http://arxiv.org/abs/1802.04715v3","url_pdf":"http://arxiv.org/pdf/1802.04715v3.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":"online-variance-reduction-for-stochastic","repo_url":"https://github.com/zalanborsos/online-variance-reduction","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}},{"paper_slug":"online-variance-reduction-for-stochastic","repo_url":"https://github.com/zalanborsos/variance-reduction-mixtures","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"stochastic-optimization","task_name":"Stochastic Optimization"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1802.04715","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}