Matches in SemOpenAlex for { <https://semopenalex.org/work/W2546700349> ?p ?o ?g. }
- W2546700349 endingPage "8" @default.
- W2546700349 startingPage "1" @default.
- W2546700349 abstract "ℓp regularization has been a popular pruning method for neural networks. The parameter p was usually set as 0<p≤2 in the literature, and practical training algorithms with ℓ0 regularization are lacking due to the NP-hard nature of the ℓ0 regularization problem; however, the ℓ0 regularization tends to produce the sparsest solution, corresponding to the most parsimonious network structure which is desirable in view of the generalization ability. To this end, this paper considers an online gradient training algorithm with smoothing ℓ0 regularization (OGTSL0) for feedforward neural networks, where the ℓ0 regularizer is approximated by a series of smoothing functions. The underlying principle for the sparsity of OGTSL0 is provided, and the convergence of the algorithm is also theoretically analyzed. Simulation examples support the theoretical analysis and illustrate the superiority of the proposed algorithm." @default.
- W2546700349 created "2016-11-11" @default.
- W2546700349 creator A5020207557 @default.
- W2546700349 creator A5070709334 @default.
- W2546700349 date "2017-02-01" @default.
- W2546700349 modified "2023-10-18" @default.
- W2546700349 title "Online gradient method with smoothing ℓ 0 regularization for feedforward neural networks" @default.
- W2546700349 cites W1969625066 @default.
- W2546700349 cites W1971942712 @default.
- W2546700349 cites W1982779742 @default.
- W2546700349 cites W2001172396 @default.
- W2546700349 cites W2004715276 @default.
- W2546700349 cites W2011873184 @default.
- W2546700349 cites W2016782026 @default.
- W2546700349 cites W2025546488 @default.
- W2546700349 cites W2025594527 @default.
- W2546700349 cites W2028069051 @default.
- W2546700349 cites W2030302676 @default.
- W2546700349 cites W2031001353 @default.
- W2546700349 cites W2048583434 @default.
- W2546700349 cites W2053795860 @default.
- W2546700349 cites W2056201402 @default.
- W2546700349 cites W2062803420 @default.
- W2546700349 cites W2070188654 @default.
- W2546700349 cites W2084130824 @default.
- W2546700349 cites W2088350412 @default.
- W2546700349 cites W2097400615 @default.
- W2546700349 cites W2097533491 @default.
- W2546700349 cites W2099005356 @default.
- W2546700349 cites W2099579348 @default.
- W2546700349 cites W2104948848 @default.
- W2546700349 cites W2104974239 @default.
- W2546700349 cites W2111072639 @default.
- W2546700349 cites W2111719156 @default.
- W2546700349 cites W2137983211 @default.
- W2546700349 cites W2138747680 @default.
- W2546700349 cites W2142635246 @default.
- W2546700349 cites W2145085734 @default.
- W2546700349 cites W2147331470 @default.
- W2546700349 cites W2153989749 @default.
- W2546700349 cites W2154987621 @default.
- W2546700349 cites W2155482699 @default.
- W2546700349 cites W2158054309 @default.
- W2546700349 cites W2168175751 @default.
- W2546700349 cites W2168728319 @default.
- W2546700349 cites W2334522350 @default.
- W2546700349 cites W2604975686 @default.
- W2546700349 doi "https://doi.org/10.1016/j.neucom.2016.10.057" @default.
- W2546700349 hasPublicationYear "2017" @default.
- W2546700349 type Work @default.
- W2546700349 sameAs 2546700349 @default.
- W2546700349 citedByCount "14" @default.
- W2546700349 countsByYear W25467003492017 @default.
- W2546700349 countsByYear W25467003492018 @default.
- W2546700349 countsByYear W25467003492019 @default.
- W2546700349 countsByYear W25467003492021 @default.
- W2546700349 countsByYear W25467003492022 @default.
- W2546700349 countsByYear W25467003492023 @default.
- W2546700349 crossrefType "journal-article" @default.
- W2546700349 hasAuthorship W2546700349A5020207557 @default.
- W2546700349 hasAuthorship W2546700349A5070709334 @default.
- W2546700349 hasConcept C11413529 @default.
- W2546700349 hasConcept C126255220 @default.
- W2546700349 hasConcept C134306372 @default.
- W2546700349 hasConcept C135252773 @default.
- W2546700349 hasConcept C141718189 @default.
- W2546700349 hasConcept C152442038 @default.
- W2546700349 hasConcept C154945302 @default.
- W2546700349 hasConcept C2776135515 @default.
- W2546700349 hasConcept C27872270 @default.
- W2546700349 hasConcept C31972630 @default.
- W2546700349 hasConcept C33923547 @default.
- W2546700349 hasConcept C3770464 @default.
- W2546700349 hasConcept C41008148 @default.
- W2546700349 hasConcept C47702885 @default.
- W2546700349 hasConcept C50644808 @default.
- W2546700349 hasConcept C79248915 @default.
- W2546700349 hasConceptScore W2546700349C11413529 @default.
- W2546700349 hasConceptScore W2546700349C126255220 @default.
- W2546700349 hasConceptScore W2546700349C134306372 @default.
- W2546700349 hasConceptScore W2546700349C135252773 @default.
- W2546700349 hasConceptScore W2546700349C141718189 @default.
- W2546700349 hasConceptScore W2546700349C152442038 @default.
- W2546700349 hasConceptScore W2546700349C154945302 @default.
- W2546700349 hasConceptScore W2546700349C2776135515 @default.
- W2546700349 hasConceptScore W2546700349C27872270 @default.
- W2546700349 hasConceptScore W2546700349C31972630 @default.
- W2546700349 hasConceptScore W2546700349C33923547 @default.
- W2546700349 hasConceptScore W2546700349C3770464 @default.
- W2546700349 hasConceptScore W2546700349C41008148 @default.
- W2546700349 hasConceptScore W2546700349C47702885 @default.
- W2546700349 hasConceptScore W2546700349C50644808 @default.
- W2546700349 hasConceptScore W2546700349C79248915 @default.
- W2546700349 hasFunder F4320321001 @default.
- W2546700349 hasFunder F4320335787 @default.
- W2546700349 hasLocation W25467003491 @default.
- W2546700349 hasOpenAccess W2546700349 @default.
- W2546700349 hasPrimaryLocation W25467003491 @default.