Matches in SemOpenAlex for { <https://semopenalex.org/work/W3123987833> ?p ?o ?g. }
Showing items 1 to 82 of
82
with 100 items per page.
- W3123987833 endingPage "173" @default.
- W3123987833 startingPage "157" @default.
- W3123987833 abstract "StochAstic Recursive grAdient algoritHm (SARAH), originally proposed for convex optimization and also proven to be effective for general nonconvex optimization, has received great attention because of its simple recursive framework for updating stochastic gradient estimates. The performance of SARAH significantly depends on the choice of the step size sequence. However, SARAH and its variants often manually select a best-tuned step size, which is time consuming in practice. Motivated by this gap, we propose a variant of the Barzilai-Borwein (BB) method, referred to as the Random Barzilai-Borwein (RBB) method, to determine the step size for SARAH in the mini-batch setting, leading to a new SARAH method: MB-SARAH-RBB. We prove that MB-SARAH-RBB converges linearly in expectation for strongly convex objective functions. Moreover, we analyze the gradient complexity of MB-SARAH-RBB and show that it is better than the original method. To further confirm the efficacy of the RBB method, we propose the MB-SARAH+-RBB method, by incorporating it into the MB-SARAH + method. Numerical experiments on standard data sets indicate that our proposed methods outperform or match state-of-the-art algorithms." @default.
- W3123987833 created "2021-02-01" @default.
- W3123987833 creator A5000518767 @default.
- W3123987833 creator A5001391971 @default.
- W3123987833 creator A5077266899 @default.
- W3123987833 date "2021-05-01" @default.
- W3123987833 modified "2023-10-02" @default.
- W3123987833 title "Accelerating mini-batch SARAH by step size rules" @default.
- W3123987833 cites W1885185971 @default.
- W3123987833 cites W1994616650 @default.
- W3123987833 cites W1996235804 @default.
- W3123987833 cites W2047152541 @default.
- W3123987833 cites W2061570747 @default.
- W3123987833 cites W2076605490 @default.
- W3123987833 cites W2080335539 @default.
- W3123987833 cites W2103628399 @default.
- W3123987833 cites W2765352990 @default.
- W3123987833 cites W2796479735 @default.
- W3123987833 cites W2810910646 @default.
- W3123987833 cites W2903514798 @default.
- W3123987833 cites W2913966198 @default.
- W3123987833 cites W2962867868 @default.
- W3123987833 cites W2963433607 @default.
- W3123987833 cites W2964185543 @default.
- W3123987833 cites W573246709 @default.
- W3123987833 doi "https://doi.org/10.1016/j.ins.2020.12.075" @default.
- W3123987833 hasPublicationYear "2021" @default.
- W3123987833 type Work @default.
- W3123987833 sameAs 3123987833 @default.
- W3123987833 citedByCount "6" @default.
- W3123987833 countsByYear W31239878332021 @default.
- W3123987833 countsByYear W31239878332022 @default.
- W3123987833 countsByYear W31239878332023 @default.
- W3123987833 crossrefType "journal-article" @default.
- W3123987833 hasAuthorship W3123987833A5000518767 @default.
- W3123987833 hasAuthorship W3123987833A5001391971 @default.
- W3123987833 hasAuthorship W3123987833A5077266899 @default.
- W3123987833 hasBestOaLocation W31239878332 @default.
- W3123987833 hasConcept C111472728 @default.
- W3123987833 hasConcept C112680207 @default.
- W3123987833 hasConcept C11413529 @default.
- W3123987833 hasConcept C115680565 @default.
- W3123987833 hasConcept C126255220 @default.
- W3123987833 hasConcept C138885662 @default.
- W3123987833 hasConcept C157972887 @default.
- W3123987833 hasConcept C2524010 @default.
- W3123987833 hasConcept C2780586882 @default.
- W3123987833 hasConcept C33923547 @default.
- W3123987833 hasConcept C41008148 @default.
- W3123987833 hasConceptScore W3123987833C111472728 @default.
- W3123987833 hasConceptScore W3123987833C112680207 @default.
- W3123987833 hasConceptScore W3123987833C11413529 @default.
- W3123987833 hasConceptScore W3123987833C115680565 @default.
- W3123987833 hasConceptScore W3123987833C126255220 @default.
- W3123987833 hasConceptScore W3123987833C138885662 @default.
- W3123987833 hasConceptScore W3123987833C157972887 @default.
- W3123987833 hasConceptScore W3123987833C2524010 @default.
- W3123987833 hasConceptScore W3123987833C2780586882 @default.
- W3123987833 hasConceptScore W3123987833C33923547 @default.
- W3123987833 hasConceptScore W3123987833C41008148 @default.
- W3123987833 hasFunder F4320321543 @default.
- W3123987833 hasLocation W31239878331 @default.
- W3123987833 hasLocation W31239878332 @default.
- W3123987833 hasOpenAccess W3123987833 @default.
- W3123987833 hasPrimaryLocation W31239878331 @default.
- W3123987833 hasRelatedWork W1887191277 @default.
- W3123987833 hasRelatedWork W2018944617 @default.
- W3123987833 hasRelatedWork W2054072800 @default.
- W3123987833 hasRelatedWork W2334954212 @default.
- W3123987833 hasRelatedWork W2589063463 @default.
- W3123987833 hasRelatedWork W2964285269 @default.
- W3123987833 hasRelatedWork W2992250528 @default.
- W3123987833 hasRelatedWork W4235288607 @default.
- W3123987833 hasRelatedWork W4248389398 @default.
- W3123987833 hasRelatedWork W4297899479 @default.
- W3123987833 hasVolume "558" @default.
- W3123987833 isParatext "false" @default.
- W3123987833 isRetracted "false" @default.
- W3123987833 magId "3123987833" @default.
- W3123987833 workType "article" @default.