{"id":9753,"date":"2026-01-15T07:02:56","date_gmt":"2026-01-15T07:02:56","guid":{"rendered":"https:\/\/mailitics.com\/index.php\/2026\/01\/15\/2601-09042\/"},"modified":"2026-01-15T07:02:56","modified_gmt":"2026-01-15T07:02:56","slug":"2601-09042","status":"publish","type":"post","link":"https:\/\/mailitics.com\/index.php\/2026\/01\/15\/2601-09042\/","title":{"rendered":"SCaLE: Switching Cost aware Learning and Exploration"},"content":{"rendered":"<p>    SCaLE: Switching Cost aware Learning and Exploration<br \/>\n \t<BR><br \/>\n<BR><\/BR><br \/>\n    <!-- no image --><br \/>\n \t<BR><br \/>\n<BR><\/BR><\/p>\n<div>arXiv:2601.09042v1 Announce Type: cross<br \/>\nAbstract: This work addresses the fundamental problem of unbounded metric movement costs in bandit online convex optimization, by considering high-dimensional dynamic quadratic hitting costs and $ell_2$-norm switching costs in a noisy bandit feedback model. For a general class of stochastic environments, we provide the first algorithm SCaLE that provably achieves a distribution-agnostic sub-linear dynamic regret, without the knowledge of hitting cost structure. En-route, we present a novel spectral regret analysis that separately quantifies eigenvalue-error driven regret and eigenbasis-perturbation driven regret. Extensive numerical experiments, against online-learning baselines, corroborate our claims, and highlight statistical consistency of our algorithm.<\/div>\n<p> \t<BR><br \/>\n <BR><\/BR><br \/>\n    Neelkamal Bhuyan, Debankur Mukherjee, Adam Wierman<br \/>\n \t<BR><br \/>\n<BR><\/BR><br \/>\n<a href=\"https:\/\/arxiv.org\/abs\/2601.09042\">Go to original source<\/a><br \/>\n \t<BR><br \/>\n <BR><\/BR><\/p>\n","protected":false},"excerpt":{"rendered":"<p>SCaLE: Switching Cost aware Learning and Exploration arXiv:2601.09042v1 Announce Type: cross Abstract: This work addresses the fundamental problem of unbounded metric movement costs in bandit online convex optimization, by considering high-dimensional dynamic quadratic hitting costs and $ell_2$-norm switching costs in a noisy bandit feedback model. For a general class of stochastic environments, we provide the [&hellip;]<\/p>\n","protected":false},"author":2,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[62,413,113,376,420,112],"tags":[660,2085,4571],"class_list":["post-9753","post","type-post","status-publish","format-standard","hentry","category-aimldsaimlds","category-cs-ds","category-cs-lg","category-math-oc","category-math-pr","category-stat-ml","tag-regret","tag-scale","tag-switching"],"_links":{"self":[{"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/posts\/9753"}],"collection":[{"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/comments?post=9753"}],"version-history":[{"count":0,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/posts\/9753\/revisions"}],"wp:attachment":[{"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/media?parent=9753"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/categories?post=9753"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/mailitics.com\/index.php\/wp-json\/wp\/v2\/tags?post=9753"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}