{"title":"弗兰克-沃尔夫算法的灵活分块迭代分析","authors":"Gábor Braun, Sebastian Pokutta, Zev Woodstock","doi":"arxiv-2409.06931","DOIUrl":null,"url":null,"abstract":"We prove that the block-coordinate Frank-Wolfe (BCFW) algorithm converges\nwith state-of-the-art rates in both convex and nonconvex settings under a very\nmild \"block-iterative\" assumption, newly allowing for (I) progress without\nactivating the most-expensive linear minimization oracle(s), LMO(s), at every\niteration, (II) parallelized updates that do not require all LMOs, and\ntherefore (III) deterministic parallel update strategies that take into account\nthe numerical cost of the problem's LMOs. Our results apply for short-step BCFW\nas well as an adaptive method for convex functions. New relationships between\nupdated coordinates and primal progress are proven, and a favorable speedup is\ndemonstrated using FrankWolfe.jl.","PeriodicalId":501286,"journal":{"name":"arXiv - MATH - Optimization and Control","volume":null,"pages":null},"PeriodicalIF":0.0000,"publicationDate":"2024-09-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Flexible block-iterative analysis for the Frank-Wolfe algorithm\",\"authors\":\"Gábor Braun, Sebastian Pokutta, Zev Woodstock\",\"doi\":\"arxiv-2409.06931\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"We prove that the block-coordinate Frank-Wolfe (BCFW) algorithm converges\\nwith state-of-the-art rates in both convex and nonconvex settings under a very\\nmild \\\"block-iterative\\\" assumption, newly allowing for (I) progress without\\nactivating the most-expensive linear minimization oracle(s), LMO(s), at every\\niteration, (II) parallelized updates that do not require all LMOs, and\\ntherefore (III) deterministic parallel update strategies that take into account\\nthe numerical cost of the problem's LMOs. Our results apply for short-step BCFW\\nas well as an adaptive method for convex functions. New relationships between\\nupdated coordinates and primal progress are proven, and a favorable speedup is\\ndemonstrated using FrankWolfe.jl.\",\"PeriodicalId\":501286,\"journal\":{\"name\":\"arXiv - MATH - Optimization and Control\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":0.0000,\"publicationDate\":\"2024-09-11\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"arXiv - MATH - Optimization and Control\",\"FirstCategoryId\":\"1085\",\"ListUrlMain\":\"https://doi.org/arxiv-2409.06931\",\"RegionNum\":0,\"RegionCategory\":null,\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"\",\"JCRName\":\"\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"arXiv - MATH - Optimization and Control","FirstCategoryId":"1085","ListUrlMain":"https://doi.org/arxiv-2409.06931","RegionNum":0,"RegionCategory":null,"ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"","JCRName":"","Score":null,"Total":0}
Flexible block-iterative analysis for the Frank-Wolfe algorithm
We prove that the block-coordinate Frank-Wolfe (BCFW) algorithm converges
with state-of-the-art rates in both convex and nonconvex settings under a very
mild "block-iterative" assumption, newly allowing for (I) progress without
activating the most-expensive linear minimization oracle(s), LMO(s), at every
iteration, (II) parallelized updates that do not require all LMOs, and
therefore (III) deterministic parallel update strategies that take into account
the numerical cost of the problem's LMOs. Our results apply for short-step BCFW
as well as an adaptive method for convex functions. New relationships between
updated coordinates and primal progress are proven, and a favorable speedup is
demonstrated using FrankWolfe.jl.