FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2912115793> ?p ?o ?g. }

• W2912115793 abstract "Stochastic gradient descent (SGD) is a popular and efficient method with wide applications in training deep neural nets and other nonconvex models. While the behavior of SGD is well understood in the convex learning setting, the existing theoretical results for SGD applied to nonconvex objective functions are far from mature. For example, existing results require to impose a nontrivial assumption on the uniform boundedness of gradients for all iterates encountered in the learning process, which is hard to verify in practical implementations. In this paper, we establish a rigorous theoretical foundation for SGD in nonconvex learning by showing that this boundedness assumption can be removed without affecting convergence rates. In particular, we establish sufficient conditions for almost sure convergence as well as optimal convergence rates for SGD applied to both general nonconvex objective functions and gradient-dominated objective functions. A linear convergence is further derived in the case with zero variances." @default.
• W2912115793 created "2019-02-21" @default.
• W2912115793 creator A5021254405 @default.
• W2912115793 creator A5046468616 @default.
• W2912115793 creator A5057631131 @default.
• W2912115793 creator A5059680084 @default.
• W2912115793 date "2019-02-03" @default.
• W2912115793 modified "2023-10-03" @default.
• W2912115793 title "Stochastic Gradient Descent for Nonconvex Learning without Bounded Gradient Assumptions" @default.
• W2912115793 cites W131378802 @default.
• W2912115793 cites W1969250928 @default.
• W2912115793 cites W1971942712 @default.
• W2912115793 cites W2016384870 @default.
• W2912115793 cites W2029463628 @default.
• W2912115793 cites W2033855314 @default.
• W2912115793 cites W2059571389 @default.
• W2912115793 cites W2087789467 @default.
• W2912115793 cites W2091825929 @default.
• W2912115793 cites W2096199223 @default.
• W2912115793 cites W2103195393 @default.
• W2912115793 cites W2112269233 @default.
• W2912115793 cites W2122678453 @default.
• W2912115793 cites W2129160848 @default.
• W2912115793 cites W2151695970 @default.
• W2912115793 cites W2334285086 @default.
• W2912115793 cites W2607907597 @default.
• W2912115793 cites W2613940844 @default.
• W2912115793 cites W2615421548 @default.
• W2912115793 cites W2741924441 @default.
• W2912115793 cites W2791759757 @default.
• W2912115793 cites W2803423166 @default.
• W2912115793 cites W2889643881 @default.
• W2912115793 cites W2952728270 @default.
• W2912115793 cites W2962876518 @default.
• W2912115793 cites W2963248893 @default.
• W2912115793 cites W2963335821 @default.
• W2912115793 cites W2963386061 @default.
• W2912115793 cites W2963433607 @default.
• W2912115793 cites W2963470657 @default.
• W2912115793 cites W2963499939 @default.
• W2912115793 cites W2963965485 @default.
• W2912115793 cites W2964120262 @default.
• W2912115793 cites W2964224116 @default.
• W2912115793 doi "https://doi.org/10.48550/arxiv.1902.00908" @default.
• W2912115793 hasPublicationYear "2019" @default.
• W2912115793 type Work @default.
• W2912115793 sameAs 2912115793 @default.
• W2912115793 citedByCount "3" @default.
• W2912115793 countsByYear W29121157932019 @default.
• W2912115793 countsByYear W29121157932020 @default.
• W2912115793 countsByYear W29121157932021 @default.
• W2912115793 crossrefType "posted-content" @default.
• W2912115793 hasAuthorship W2912115793A5021254405 @default.
• W2912115793 hasAuthorship W2912115793A5046468616 @default.
• W2912115793 hasAuthorship W2912115793A5057631131 @default.
• W2912115793 hasAuthorship W2912115793A5059680084 @default.
• W2912115793 hasBestOaLocation W29121157931 @default.
• W2912115793 hasConcept C111919701 @default.
• W2912115793 hasConcept C112680207 @default.
• W2912115793 hasConcept C126255220 @default.
• W2912115793 hasConcept C134306372 @default.
• W2912115793 hasConcept C140479938 @default.
• W2912115793 hasConcept C145446738 @default.
• W2912115793 hasConcept C153258448 @default.
• W2912115793 hasConcept C154945302 @default.
• W2912115793 hasConcept C162324750 @default.
• W2912115793 hasConcept C206688291 @default.
• W2912115793 hasConcept C2524010 @default.
• W2912115793 hasConcept C2777303404 @default.
• W2912115793 hasConcept C28826006 @default.
• W2912115793 hasConcept C33923547 @default.
• W2912115793 hasConcept C34388435 @default.
• W2912115793 hasConcept C41008148 @default.
• W2912115793 hasConcept C50522688 @default.
• W2912115793 hasConcept C50644808 @default.
• W2912115793 hasConcept C98045186 @default.
• W2912115793 hasConceptScore W2912115793C111919701 @default.
• W2912115793 hasConceptScore W2912115793C112680207 @default.
• W2912115793 hasConceptScore W2912115793C126255220 @default.
• W2912115793 hasConceptScore W2912115793C134306372 @default.
• W2912115793 hasConceptScore W2912115793C140479938 @default.
• W2912115793 hasConceptScore W2912115793C145446738 @default.
• W2912115793 hasConceptScore W2912115793C153258448 @default.
• W2912115793 hasConceptScore W2912115793C154945302 @default.
• W2912115793 hasConceptScore W2912115793C162324750 @default.
• W2912115793 hasConceptScore W2912115793C206688291 @default.
• W2912115793 hasConceptScore W2912115793C2524010 @default.
• W2912115793 hasConceptScore W2912115793C2777303404 @default.
• W2912115793 hasConceptScore W2912115793C28826006 @default.
• W2912115793 hasConceptScore W2912115793C33923547 @default.
• W2912115793 hasConceptScore W2912115793C34388435 @default.
• W2912115793 hasConceptScore W2912115793C41008148 @default.
• W2912115793 hasConceptScore W2912115793C50522688 @default.
• W2912115793 hasConceptScore W2912115793C50644808 @default.
• W2912115793 hasConceptScore W2912115793C98045186 @default.
• W2912115793 hasLocation W29121157931 @default.
• W2912115793 hasOpenAccess W2912115793 @default.
• W2912115793 hasPrimaryLocation W29121157931 @default.
• W2912115793 hasRelatedWork W2338191349 @default.
• W2912115793 hasRelatedWork W2523772899 @default.

• next