Matches in SemOpenAlex for { <https://semopenalex.org/work/W4297797483> ?p ?o ?g. }
Showing items 1 to 61 of
61
with 100 items per page.
- W4297797483 abstract "Over the past ten years, driven by large scale optimisation problems arising from machine learning, the development of stochastic optimisation methods have witnessed a tremendous growth. However, despite their popularity, the theoretical understandings of these methods are quite limited in contrast to the deterministic optimisation methods. In this paper, we present a local convergence analysis for a typical type of stochastic optimisation methods: proximal variance reduced stochastic gradient methods, and mainly focus on the SAGA [12] and Prox-SVRG [43] algorithms. Under the assumption that the non-smooth component of the optimisation problem is partly smooth relative to a smooth manifold, we present a unified framework for the local convergence analysis of the SAGA/Prox-SVRG algorithms: (i) the sequences generated by the SAGA/Prox-SVRG are able to identify the smooth manifold in a finite number of iterations; (ii) then the sequence enters a local linear convergence regime. Beyond local convergence analysis, we also discuss various possibilities for accelerating these algorithms, including adapting to better local parameters, and applying higher-order deterministic/stochastic optimisation methods which can achieve super-linear convergence. Concrete examples arising from machine learning are considered to verify the obtained results." @default.
- W4297797483 created "2022-10-01" @default.
- W4297797483 creator A5004350216 @default.
- W4297797483 creator A5033880300 @default.
- W4297797483 creator A5066789410 @default.
- W4297797483 date "2018-02-07" @default.
- W4297797483 modified "2023-10-18" @default.
- W4297797483 title "Local Convergence Properties of SAGA/Prox-SVRG and Acceleration" @default.
- W4297797483 doi "https://doi.org/10.48550/arxiv.1802.02554" @default.
- W4297797483 hasPublicationYear "2018" @default.
- W4297797483 type Work @default.
- W4297797483 citedByCount "0" @default.
- W4297797483 crossrefType "posted-content" @default.
- W4297797483 hasAuthorship W4297797483A5004350216 @default.
- W4297797483 hasAuthorship W4297797483A5033880300 @default.
- W4297797483 hasAuthorship W4297797483A5066789410 @default.
- W4297797483 hasBestOaLocation W42977974831 @default.
- W4297797483 hasConcept C11413529 @default.
- W4297797483 hasConcept C117896860 @default.
- W4297797483 hasConcept C121332964 @default.
- W4297797483 hasConcept C126255220 @default.
- W4297797483 hasConcept C127413603 @default.
- W4297797483 hasConcept C154945302 @default.
- W4297797483 hasConcept C162324750 @default.
- W4297797483 hasConcept C2777303404 @default.
- W4297797483 hasConcept C33923547 @default.
- W4297797483 hasConcept C41008148 @default.
- W4297797483 hasConcept C50522688 @default.
- W4297797483 hasConcept C529865628 @default.
- W4297797483 hasConcept C74650414 @default.
- W4297797483 hasConcept C78519656 @default.
- W4297797483 hasConceptScore W4297797483C11413529 @default.
- W4297797483 hasConceptScore W4297797483C117896860 @default.
- W4297797483 hasConceptScore W4297797483C121332964 @default.
- W4297797483 hasConceptScore W4297797483C126255220 @default.
- W4297797483 hasConceptScore W4297797483C127413603 @default.
- W4297797483 hasConceptScore W4297797483C154945302 @default.
- W4297797483 hasConceptScore W4297797483C162324750 @default.
- W4297797483 hasConceptScore W4297797483C2777303404 @default.
- W4297797483 hasConceptScore W4297797483C33923547 @default.
- W4297797483 hasConceptScore W4297797483C41008148 @default.
- W4297797483 hasConceptScore W4297797483C50522688 @default.
- W4297797483 hasConceptScore W4297797483C529865628 @default.
- W4297797483 hasConceptScore W4297797483C74650414 @default.
- W4297797483 hasConceptScore W4297797483C78519656 @default.
- W4297797483 hasLocation W42977974831 @default.
- W4297797483 hasOpenAccess W4297797483 @default.
- W4297797483 hasPrimaryLocation W42977974831 @default.
- W4297797483 hasRelatedWork W1506216622 @default.
- W4297797483 hasRelatedWork W1987033902 @default.
- W4297797483 hasRelatedWork W2084146399 @default.
- W4297797483 hasRelatedWork W2090128564 @default.
- W4297797483 hasRelatedWork W2152713700 @default.
- W4297797483 hasRelatedWork W2322296892 @default.
- W4297797483 hasRelatedWork W2374839191 @default.
- W4297797483 hasRelatedWork W2980477753 @default.
- W4297797483 hasRelatedWork W3034926706 @default.
- W4297797483 hasRelatedWork W3107474891 @default.
- W4297797483 isParatext "false" @default.
- W4297797483 isRetracted "false" @default.
- W4297797483 workType "article" @default.