Matches in SemOpenAlex for { <https://semopenalex.org/work/W3008480154> ?p ?o ?g. }
Showing items 1 to 90 of
90
with 100 items per page.
- W3008480154 endingPage "289" @default.
- W3008480154 startingPage "281" @default.
- W3008480154 abstract "Rectified activation units make an important contribution to the success of deep neural networks in many computer vision tasks. In this paper, we propose a Parametric Deformable Exponential Linear Unit (PDELU) and theoretically verify its effectiveness for improving the convergence speed of learning procedure. By means of flexible map shape, the proposed PDELU could push the mean value of activation responses closer to zero, which ensures the steepest descent in training a deep neural network. We verify the effectiveness of the proposed method in the image classification task. Extensive experiments on three classical databases (i.e., CIFAR-10, CIFAR-100, and ImageNet-2015) indicate that the proposed method leads to higher convergence speed and better accuracy when it is embedded into different CNN architectures (i.e., NIN, ResNet, WRN, and DenseNet). Meanwhile, the proposed PDELU outperforms many existing shape-specific activation functions (i.e., Maxout, ReLU, LeakyReLU, ELU, SELU, SoftPlus, Swish) and the shape-adaptive activation functions (i.e., APL, PReLU, MPELU, FReLU)." @default.
- W3008480154 created "2020-03-06" @default.
- W3008480154 creator A5022134460 @default.
- W3008480154 creator A5075737786 @default.
- W3008480154 creator A5077165659 @default.
- W3008480154 creator A5084658549 @default.
- W3008480154 creator A5087761862 @default.
- W3008480154 date "2020-05-01" @default.
- W3008480154 modified "2023-10-15" @default.
- W3008480154 title "Parametric Deformable Exponential Linear Units for deep neural networks" @default.
- W3008480154 cites W2094498958 @default.
- W3008480154 cites W2117539524 @default.
- W3008480154 cites W2503931548 @default.
- W3008480154 cites W2964055987 @default.
- W3008480154 cites W3100954504 @default.
- W3008480154 cites W4241400578 @default.
- W3008480154 doi "https://doi.org/10.1016/j.neunet.2020.02.012" @default.
- W3008480154 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/32151915" @default.
- W3008480154 hasPublicationYear "2020" @default.
- W3008480154 type Work @default.
- W3008480154 sameAs 3008480154 @default.
- W3008480154 citedByCount "22" @default.
- W3008480154 countsByYear W30084801542020 @default.
- W3008480154 countsByYear W30084801542021 @default.
- W3008480154 countsByYear W30084801542022 @default.
- W3008480154 countsByYear W30084801542023 @default.
- W3008480154 crossrefType "journal-article" @default.
- W3008480154 hasAuthorship W3008480154A5022134460 @default.
- W3008480154 hasAuthorship W3008480154A5075737786 @default.
- W3008480154 hasAuthorship W3008480154A5077165659 @default.
- W3008480154 hasAuthorship W3008480154A5084658549 @default.
- W3008480154 hasAuthorship W3008480154A5087761862 @default.
- W3008480154 hasConcept C105795698 @default.
- W3008480154 hasConcept C108583219 @default.
- W3008480154 hasConcept C11413529 @default.
- W3008480154 hasConcept C117251300 @default.
- W3008480154 hasConcept C153180895 @default.
- W3008480154 hasConcept C153258448 @default.
- W3008480154 hasConcept C154945302 @default.
- W3008480154 hasConcept C162324750 @default.
- W3008480154 hasConcept C2524010 @default.
- W3008480154 hasConcept C2777303404 @default.
- W3008480154 hasConcept C2944601119 @default.
- W3008480154 hasConcept C2984842247 @default.
- W3008480154 hasConcept C33923547 @default.
- W3008480154 hasConcept C38365724 @default.
- W3008480154 hasConcept C41008148 @default.
- W3008480154 hasConcept C50522688 @default.
- W3008480154 hasConcept C50644808 @default.
- W3008480154 hasConcept C99636146 @default.
- W3008480154 hasConceptScore W3008480154C105795698 @default.
- W3008480154 hasConceptScore W3008480154C108583219 @default.
- W3008480154 hasConceptScore W3008480154C11413529 @default.
- W3008480154 hasConceptScore W3008480154C117251300 @default.
- W3008480154 hasConceptScore W3008480154C153180895 @default.
- W3008480154 hasConceptScore W3008480154C153258448 @default.
- W3008480154 hasConceptScore W3008480154C154945302 @default.
- W3008480154 hasConceptScore W3008480154C162324750 @default.
- W3008480154 hasConceptScore W3008480154C2524010 @default.
- W3008480154 hasConceptScore W3008480154C2777303404 @default.
- W3008480154 hasConceptScore W3008480154C2944601119 @default.
- W3008480154 hasConceptScore W3008480154C2984842247 @default.
- W3008480154 hasConceptScore W3008480154C33923547 @default.
- W3008480154 hasConceptScore W3008480154C38365724 @default.
- W3008480154 hasConceptScore W3008480154C41008148 @default.
- W3008480154 hasConceptScore W3008480154C50522688 @default.
- W3008480154 hasConceptScore W3008480154C50644808 @default.
- W3008480154 hasConceptScore W3008480154C99636146 @default.
- W3008480154 hasFunder F4320321001 @default.
- W3008480154 hasLocation W30084801541 @default.
- W3008480154 hasLocation W30084801542 @default.
- W3008480154 hasOpenAccess W3008480154 @default.
- W3008480154 hasPrimaryLocation W30084801541 @default.
- W3008480154 hasRelatedWork W2082482750 @default.
- W3008480154 hasRelatedWork W2279398222 @default.
- W3008480154 hasRelatedWork W2544122622 @default.
- W3008480154 hasRelatedWork W2738221750 @default.
- W3008480154 hasRelatedWork W2767072113 @default.
- W3008480154 hasRelatedWork W2791691546 @default.
- W3008480154 hasRelatedWork W2799384463 @default.
- W3008480154 hasRelatedWork W2915754718 @default.
- W3008480154 hasRelatedWork W3008480154 @default.
- W3008480154 hasRelatedWork W4299822940 @default.
- W3008480154 hasVolume "125" @default.
- W3008480154 isParatext "false" @default.
- W3008480154 isRetracted "false" @default.
- W3008480154 magId "3008480154" @default.
- W3008480154 workType "article" @default.