Matches in SemOpenAlex for { <https://semopenalex.org/work/W3107422805> ?p ?o ?g. }
Showing items 1 to 76 of
76
with 100 items per page.
- W3107422805 endingPage "106626" @default.
- W3107422805 startingPage "106626" @default.
- W3107422805 abstract "Abstract The question of how to readily reduce variance in stochastic optimization methods is challenging. The most widely used approach of reducing variance in stochastic optimization is to employ past gradients. Although utilizing second order information in stochastic optimization has recently been proved to be an effective way to track gradients, research on taking second order information to track gradients is quite limited due to a high computational burden. In this paper, we introduce the modified form of the implicit gradient transport (MIGT) technique into the classical stochastic variance reduced methods, i.e., SVRG, which leads to a new method called SVRG–MIGT. The SVRG–MIGT method enjoys second order information to provide faster variance reduction, but without computing the Hessian explicitly. We analyze the theoretical convergence properties of the SVRG–MIGT method for strongly convex objective functions and prove that it is comparable to state-of-the-art methods. To further confirm the effectiveness of MIGT, we propose Acc-Prox-SVRG–MIGT, a novel variant of the accelerated mini-batch Prox-SVRG (Acc-Prox-SVRG) method, by introducing it into Acc-Prox-SVRG. In addition, to greatly reduce the computational complexity, we propose L-SVRG–MIGT, by introducing MIGT to loopless SVRG (L-SVRG), which uses the full gradient with a small probability. Moreover, we establish a convergence analysis for L-SVRG–MIGT. Experimental results on a wide range of data sets demonstrate the efficacy and robustness of our proposed methods." @default.
- W3107422805 created "2020-12-07" @default.
- W3107422805 creator A5000518767 @default.
- W3107422805 date "2021-01-01" @default.
- W3107422805 modified "2023-09-27" @default.
- W3107422805 title "Variance reduced optimization with implicit gradient transport" @default.
- W3107422805 cites W1988720110 @default.
- W3107422805 cites W1988795359 @default.
- W3107422805 cites W1992208280 @default.
- W3107422805 cites W2055117001 @default.
- W3107422805 cites W2061570747 @default.
- W3107422805 cites W2072327470 @default.
- W3107422805 cites W2086161653 @default.
- W3107422805 cites W2100556411 @default.
- W3107422805 cites W2103628399 @default.
- W3107422805 cites W2113824004 @default.
- W3107422805 cites W2125993116 @default.
- W3107422805 cites W2147950119 @default.
- W3107422805 cites W2157011125 @default.
- W3107422805 cites W2167090712 @default.
- W3107422805 cites W2786021835 @default.
- W3107422805 cites W2796479735 @default.
- W3107422805 cites W2897969205 @default.
- W3107422805 cites W2903514798 @default.
- W3107422805 cites W2913966198 @default.
- W3107422805 cites W2919115771 @default.
- W3107422805 cites W2956353133 @default.
- W3107422805 cites W2962867868 @default.
- W3107422805 cites W2963470657 @default.
- W3107422805 cites W2963954797 @default.
- W3107422805 cites W3007343725 @default.
- W3107422805 cites W3018951082 @default.
- W3107422805 cites W4206742934 @default.
- W3107422805 doi "https://doi.org/10.1016/j.knosys.2020.106626" @default.
- W3107422805 hasPublicationYear "2021" @default.
- W3107422805 type Work @default.
- W3107422805 sameAs 3107422805 @default.
- W3107422805 citedByCount "2" @default.
- W3107422805 countsByYear W31074228052022 @default.
- W3107422805 crossrefType "journal-article" @default.
- W3107422805 hasAuthorship W3107422805A5000518767 @default.
- W3107422805 hasConcept C11413529 @default.
- W3107422805 hasConcept C121955636 @default.
- W3107422805 hasConcept C126255220 @default.
- W3107422805 hasConcept C144133560 @default.
- W3107422805 hasConcept C196083921 @default.
- W3107422805 hasConcept C33923547 @default.
- W3107422805 hasConcept C41008148 @default.
- W3107422805 hasConceptScore W3107422805C11413529 @default.
- W3107422805 hasConceptScore W3107422805C121955636 @default.
- W3107422805 hasConceptScore W3107422805C126255220 @default.
- W3107422805 hasConceptScore W3107422805C144133560 @default.
- W3107422805 hasConceptScore W3107422805C196083921 @default.
- W3107422805 hasConceptScore W3107422805C33923547 @default.
- W3107422805 hasConceptScore W3107422805C41008148 @default.
- W3107422805 hasFunder F4320321543 @default.
- W3107422805 hasLocation W31074228051 @default.
- W3107422805 hasOpenAccess W3107422805 @default.
- W3107422805 hasPrimaryLocation W31074228051 @default.
- W3107422805 hasRelatedWork W1978198811 @default.
- W3107422805 hasRelatedWork W2040505576 @default.
- W3107422805 hasRelatedWork W2061722754 @default.
- W3107422805 hasRelatedWork W2351491280 @default.
- W3107422805 hasRelatedWork W2366079397 @default.
- W3107422805 hasRelatedWork W2371447506 @default.
- W3107422805 hasRelatedWork W2377321178 @default.
- W3107422805 hasRelatedWork W2386767533 @default.
- W3107422805 hasRelatedWork W303980170 @default.
- W3107422805 hasRelatedWork W2117492019 @default.
- W3107422805 hasVolume "212" @default.
- W3107422805 isParatext "false" @default.
- W3107422805 isRetracted "false" @default.
- W3107422805 magId "3107422805" @default.
- W3107422805 workType "article" @default.