Matches in SemOpenAlex for { <https://semopenalex.org/work/W2971452831> ?p ?o ?g. }
- W2971452831 endingPage "12" @default.
- W2971452831 startingPage "1" @default.
- W2971452831 abstract "The extrapolation strategy raised by Nesterov, which can accelerate the convergence rate of gradient descent methods by orders of magnitude when dealing with smooth convex objective, has led to tremendous success in training machine learning tasks. In this article, the convergence of individual iterates of projected subgradient (PSG) methods for nonsmooth convex optimization problems is theoretically studied based on Nesterov's extrapolation, which we name individual convergence. We prove that Nesterov's extrapolation has the strength to make the individual convergence of PSG optimal for nonsmooth problems. In light of this consideration, a direct modification of the subgradient evaluation suffices to achieve optimal individual convergence for strongly convex problems, which can be regarded as making an interesting step toward the open question about stochastic gradient descent (SGD) posed by Shamir. Furthermore, we give an extension of the derived algorithms to solve regularized learning tasks with nonsmooth losses in stochastic settings. Compared with other state-of-the-art nonsmooth methods, the derived algorithms can serve as an alternative to the basic SGD especially in coping with machine learning problems, where an individual output is needed to guarantee the regularization structure while keeping an optimal rate of convergence. Typically, our method is applicable as an efficient tool for solving large-scale $l$1-regularized hinge-loss learning problems. Several comparison experiments demonstrate that our individual output not only achieves an optimal convergence rate but also guarantees better sparsity than the averaged solution." @default.
- W2971452831 created "2019-09-12" @default.
- W2971452831 creator A5037669752 @default.
- W2971452831 creator A5046355512 @default.
- W2971452831 creator A5052908126 @default.
- W2971452831 creator A5053102123 @default.
- W2971452831 date "2019-01-01" @default.
- W2971452831 modified "2023-10-16" @default.
- W2971452831 title "The Strength of Nesterov's Extrapolation in the Individual Convergence of Nonsmooth Optimization" @default.
- W2971452831 cites W2024484010 @default.
- W2971452831 cites W2028148084 @default.
- W2971452831 cites W2069170077 @default.
- W2971452831 cites W2090852255 @default.
- W2971452831 cites W2091825929 @default.
- W2971452831 cites W2100556411 @default.
- W2971452831 cites W2101868363 @default.
- W2971452831 cites W2124541940 @default.
- W2971452831 cites W2124768887 @default.
- W2971452831 cites W2129160848 @default.
- W2971452831 cites W2142623206 @default.
- W2971452831 cites W2167732364 @default.
- W2971452831 cites W2217490359 @default.
- W2971452831 cites W2413731729 @default.
- W2971452831 cites W2805747803 @default.
- W2971452831 cites W2896146108 @default.
- W2971452831 cites W2912848192 @default.
- W2971452831 cites W2963433607 @default.
- W2971452831 doi "https://doi.org/10.1109/tnnls.2019.2933452" @default.
- W2971452831 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/31484139" @default.
- W2971452831 hasPublicationYear "2019" @default.
- W2971452831 type Work @default.
- W2971452831 sameAs 2971452831 @default.
- W2971452831 citedByCount "8" @default.
- W2971452831 countsByYear W29714528312020 @default.
- W2971452831 countsByYear W29714528312021 @default.
- W2971452831 countsByYear W29714528312022 @default.
- W2971452831 countsByYear W29714528312023 @default.
- W2971452831 crossrefType "journal-article" @default.
- W2971452831 hasAuthorship W2971452831A5037669752 @default.
- W2971452831 hasAuthorship W2971452831A5046355512 @default.
- W2971452831 hasAuthorship W2971452831A5052908126 @default.
- W2971452831 hasAuthorship W2971452831A5053102123 @default.
- W2971452831 hasBestOaLocation W29714528312 @default.
- W2971452831 hasConcept C112680207 @default.
- W2971452831 hasConcept C126255220 @default.
- W2971452831 hasConcept C127162648 @default.
- W2971452831 hasConcept C132459708 @default.
- W2971452831 hasConcept C134306372 @default.
- W2971452831 hasConcept C140479938 @default.
- W2971452831 hasConcept C145446738 @default.
- W2971452831 hasConcept C154945302 @default.
- W2971452831 hasConcept C157972887 @default.
- W2971452831 hasConcept C158968445 @default.
- W2971452831 hasConcept C162324750 @default.
- W2971452831 hasConcept C206688291 @default.
- W2971452831 hasConcept C2524010 @default.
- W2971452831 hasConcept C2776135515 @default.
- W2971452831 hasConcept C2777303404 @default.
- W2971452831 hasConcept C28826006 @default.
- W2971452831 hasConcept C31258907 @default.
- W2971452831 hasConcept C33923547 @default.
- W2971452831 hasConcept C41008148 @default.
- W2971452831 hasConcept C50522688 @default.
- W2971452831 hasConcept C50644808 @default.
- W2971452831 hasConcept C57869625 @default.
- W2971452831 hasConceptScore W2971452831C112680207 @default.
- W2971452831 hasConceptScore W2971452831C126255220 @default.
- W2971452831 hasConceptScore W2971452831C127162648 @default.
- W2971452831 hasConceptScore W2971452831C132459708 @default.
- W2971452831 hasConceptScore W2971452831C134306372 @default.
- W2971452831 hasConceptScore W2971452831C140479938 @default.
- W2971452831 hasConceptScore W2971452831C145446738 @default.
- W2971452831 hasConceptScore W2971452831C154945302 @default.
- W2971452831 hasConceptScore W2971452831C157972887 @default.
- W2971452831 hasConceptScore W2971452831C158968445 @default.
- W2971452831 hasConceptScore W2971452831C162324750 @default.
- W2971452831 hasConceptScore W2971452831C206688291 @default.
- W2971452831 hasConceptScore W2971452831C2524010 @default.
- W2971452831 hasConceptScore W2971452831C2776135515 @default.
- W2971452831 hasConceptScore W2971452831C2777303404 @default.
- W2971452831 hasConceptScore W2971452831C28826006 @default.
- W2971452831 hasConceptScore W2971452831C31258907 @default.
- W2971452831 hasConceptScore W2971452831C33923547 @default.
- W2971452831 hasConceptScore W2971452831C41008148 @default.
- W2971452831 hasConceptScore W2971452831C50522688 @default.
- W2971452831 hasConceptScore W2971452831C50644808 @default.
- W2971452831 hasConceptScore W2971452831C57869625 @default.
- W2971452831 hasFunder F4320321001 @default.
- W2971452831 hasFunder F4320335777 @default.
- W2971452831 hasLocation W29714528311 @default.
- W2971452831 hasLocation W29714528312 @default.
- W2971452831 hasLocation W29714528313 @default.
- W2971452831 hasOpenAccess W2971452831 @default.
- W2971452831 hasPrimaryLocation W29714528311 @default.
- W2971452831 hasRelatedWork W1979007968 @default.
- W2971452831 hasRelatedWork W2000431671 @default.
- W2971452831 hasRelatedWork W2128927076 @default.
- W2971452831 hasRelatedWork W2505972737 @default.