Matches in SemOpenAlex for { <https://semopenalex.org/work/W4384613979> ?p ?o ?g. }
Showing items 1 to 65 of
65
with 100 items per page.
- W4384613979 abstract "In this paper, we consider the decentralized, stochastic nonconvex strongly-concave (NCSC) minimax problem with nonsmooth regularization terms on both primal and dual variables, wherein a network of $m$ computing agents collaborate via peer-to-peer communications. We consider when the coupling function is in expectation or finite-sum form and the double regularizers are convex functions, applied separately to the primal and dual variables. Our algorithmic framework introduces a Lagrangian multiplier to eliminate the consensus constraint on the dual variable. Coupling this with variance-reduction (VR) techniques, our proposed method, entitled VRLM, by a single neighbor communication per iteration, is able to achieve an $mathcal{O}(kappa^3varepsilon^{-3})$ sample complexity under the general stochastic setting, with either a big-batch or small-batch VR option, where $kappa$ is the condition number of the problem and $varepsilon$ is the desired solution accuracy. With a big-batch VR, we can additionally achieve $mathcal{O}(kappa^2varepsilon^{-2})$ communication complexity. Under the special finite-sum setting, our method with a big-batch VR can achieve an $mathcal{O}(n + sqrt{n} kappa^2varepsilon^{-2})$ sample complexity and $mathcal{O}(kappa^2varepsilon^{-2})$ communication complexity, where $n$ is the number of components in the finite sum. All complexity results match the best-known results achieved by a few existing methods for solving special cases of the problem we consider. To the best of our knowledge, this is the first work which provides convergence guarantees for NCSC minimax problems with general convex nonsmooth regularizers applied to both the primal and dual variables in the decentralized stochastic setting. Numerical experiments are conducted on two machine learning problems. Our code is downloadable from https://github.com/RPI-OPT/VRLM." @default.
- W4384613979 created "2023-07-18" @default.
- W4384613979 creator A5055895298 @default.
- W4384613979 creator A5072186400 @default.
- W4384613979 date "2023-07-13" @default.
- W4384613979 modified "2023-09-23" @default.
- W4384613979 title "Variance-reduced accelerated methods for decentralized stochastic double-regularized nonconvex strongly-concave minimax problems" @default.
- W4384613979 doi "https://doi.org/10.48550/arxiv.2307.07113" @default.
- W4384613979 hasPublicationYear "2023" @default.
- W4384613979 type Work @default.
- W4384613979 citedByCount "0" @default.
- W4384613979 crossrefType "posted-content" @default.
- W4384613979 hasAuthorship W4384613979A5055895298 @default.
- W4384613979 hasAuthorship W4384613979A5072186400 @default.
- W4384613979 hasBestOaLocation W43846139791 @default.
- W4384613979 hasConcept C105795698 @default.
- W4384613979 hasConcept C112680207 @default.
- W4384613979 hasConcept C114614502 @default.
- W4384613979 hasConcept C118615104 @default.
- W4384613979 hasConcept C126255220 @default.
- W4384613979 hasConcept C149728462 @default.
- W4384613979 hasConcept C154945302 @default.
- W4384613979 hasConcept C162324750 @default.
- W4384613979 hasConcept C19499675 @default.
- W4384613979 hasConcept C2524010 @default.
- W4384613979 hasConcept C2776135515 @default.
- W4384613979 hasConcept C2777303404 @default.
- W4384613979 hasConcept C28826006 @default.
- W4384613979 hasConcept C33923547 @default.
- W4384613979 hasConcept C41008148 @default.
- W4384613979 hasConcept C50522688 @default.
- W4384613979 hasConcept C62644790 @default.
- W4384613979 hasConceptScore W4384613979C105795698 @default.
- W4384613979 hasConceptScore W4384613979C112680207 @default.
- W4384613979 hasConceptScore W4384613979C114614502 @default.
- W4384613979 hasConceptScore W4384613979C118615104 @default.
- W4384613979 hasConceptScore W4384613979C126255220 @default.
- W4384613979 hasConceptScore W4384613979C149728462 @default.
- W4384613979 hasConceptScore W4384613979C154945302 @default.
- W4384613979 hasConceptScore W4384613979C162324750 @default.
- W4384613979 hasConceptScore W4384613979C19499675 @default.
- W4384613979 hasConceptScore W4384613979C2524010 @default.
- W4384613979 hasConceptScore W4384613979C2776135515 @default.
- W4384613979 hasConceptScore W4384613979C2777303404 @default.
- W4384613979 hasConceptScore W4384613979C28826006 @default.
- W4384613979 hasConceptScore W4384613979C33923547 @default.
- W4384613979 hasConceptScore W4384613979C41008148 @default.
- W4384613979 hasConceptScore W4384613979C50522688 @default.
- W4384613979 hasConceptScore W4384613979C62644790 @default.
- W4384613979 hasLocation W43846139791 @default.
- W4384613979 hasOpenAccess W4384613979 @default.
- W4384613979 hasPrimaryLocation W43846139791 @default.
- W4384613979 hasRelatedWork W1977563107 @default.
- W4384613979 hasRelatedWork W2023578832 @default.
- W4384613979 hasRelatedWork W2055591376 @default.
- W4384613979 hasRelatedWork W2347593305 @default.
- W4384613979 hasRelatedWork W2349727079 @default.
- W4384613979 hasRelatedWork W2371360614 @default.
- W4384613979 hasRelatedWork W3099694557 @default.
- W4384613979 hasRelatedWork W4312397237 @default.
- W4384613979 hasRelatedWork W4321854672 @default.
- W4384613979 hasRelatedWork W3200260526 @default.
- W4384613979 isParatext "false" @default.
- W4384613979 isRetracted "false" @default.
- W4384613979 workType "article" @default.