Matches in SemOpenAlex for { <https://semopenalex.org/work/W3199280775> ?p ?o ?g. }
Showing items 1 to 79 of
79
with 100 items per page.
- W3199280775 abstract "Channel pruning is a commonly used model compression in convolutional neural network. The structured pruning using sparse constraints can automatically learn the importance of parameters during the training process by imposing sparse constraints on parameters. However, existing pruning methods based on sparse constraints cannot process the final convolutional layer of the residual module with complex connections. Due to the existence of residual connection, if the final convolutional layer of the residual module is pruned, the sparse channel of the feature map from residual connection does not correspond to the feature map from module output, which will cause the parameters to be unable to be pruned. This paper studies this problem and proposes an identity association group pruning algorithm, which we call IGP. IGP groups the parameters and channels that generate the corresponding feature maps, uses Group Lasso to sparse the same group of parameters as a whole, and forces the sparseness of the parameters with sparse correlation to be consistent with each other. Experiments show that when IGP compresses ResNet56 60% parameters, the model performance only drops 0.36 %, which is better than the existing pruning method based on sparse constraints. In the case of high compression ratio, IGP can compresses ResNet-50 compressesed with 87% parameters and the performance drops only 0.76%, which is 5.17 % higher than the existing methods." @default.
- W3199280775 created "2021-09-27" @default.
- W3199280775 creator A5011489176 @default.
- W3199280775 creator A5029803377 @default.
- W3199280775 creator A5036299430 @default.
- W3199280775 creator A5037471432 @default.
- W3199280775 creator A5048111142 @default.
- W3199280775 creator A5085352453 @default.
- W3199280775 date "2021-07-18" @default.
- W3199280775 modified "2023-10-16" @default.
- W3199280775 title "Identity-linked Group Channel Pruning for Deep Neural Networks" @default.
- W3199280775 cites W2194775991 @default.
- W3199280775 cites W2276892413 @default.
- W3199280775 cites W2531409750 @default.
- W3199280775 cites W2549139847 @default.
- W3199280775 cites W2752782242 @default.
- W3199280775 cites W2962851801 @default.
- W3199280775 cites W2963363373 @default.
- W3199280775 cites W2963446712 @default.
- W3199280775 cites W2964233199 @default.
- W3199280775 doi "https://doi.org/10.1109/ijcnn52387.2021.9533908" @default.
- W3199280775 hasPublicationYear "2021" @default.
- W3199280775 type Work @default.
- W3199280775 sameAs 3199280775 @default.
- W3199280775 citedByCount "1" @default.
- W3199280775 countsByYear W31992807752023 @default.
- W3199280775 crossrefType "proceedings-article" @default.
- W3199280775 hasAuthorship W3199280775A5011489176 @default.
- W3199280775 hasAuthorship W3199280775A5029803377 @default.
- W3199280775 hasAuthorship W3199280775A5036299430 @default.
- W3199280775 hasAuthorship W3199280775A5037471432 @default.
- W3199280775 hasAuthorship W3199280775A5048111142 @default.
- W3199280775 hasAuthorship W3199280775A5085352453 @default.
- W3199280775 hasConcept C108010975 @default.
- W3199280775 hasConcept C11413529 @default.
- W3199280775 hasConcept C127162648 @default.
- W3199280775 hasConcept C138885662 @default.
- W3199280775 hasConcept C153180895 @default.
- W3199280775 hasConcept C154945302 @default.
- W3199280775 hasConcept C155512373 @default.
- W3199280775 hasConcept C2776401178 @default.
- W3199280775 hasConcept C31258907 @default.
- W3199280775 hasConcept C41008148 @default.
- W3199280775 hasConcept C41895202 @default.
- W3199280775 hasConcept C6557445 @default.
- W3199280775 hasConcept C81363708 @default.
- W3199280775 hasConcept C86803240 @default.
- W3199280775 hasConceptScore W3199280775C108010975 @default.
- W3199280775 hasConceptScore W3199280775C11413529 @default.
- W3199280775 hasConceptScore W3199280775C127162648 @default.
- W3199280775 hasConceptScore W3199280775C138885662 @default.
- W3199280775 hasConceptScore W3199280775C153180895 @default.
- W3199280775 hasConceptScore W3199280775C154945302 @default.
- W3199280775 hasConceptScore W3199280775C155512373 @default.
- W3199280775 hasConceptScore W3199280775C2776401178 @default.
- W3199280775 hasConceptScore W3199280775C31258907 @default.
- W3199280775 hasConceptScore W3199280775C41008148 @default.
- W3199280775 hasConceptScore W3199280775C41895202 @default.
- W3199280775 hasConceptScore W3199280775C6557445 @default.
- W3199280775 hasConceptScore W3199280775C81363708 @default.
- W3199280775 hasConceptScore W3199280775C86803240 @default.
- W3199280775 hasFunder F4320321001 @default.
- W3199280775 hasLocation W31992807751 @default.
- W3199280775 hasOpenAccess W3199280775 @default.
- W3199280775 hasPrimaryLocation W31992807751 @default.
- W3199280775 hasRelatedWork W2295021132 @default.
- W3199280775 hasRelatedWork W2765989371 @default.
- W3199280775 hasRelatedWork W2767090503 @default.
- W3199280775 hasRelatedWork W2767651786 @default.
- W3199280775 hasRelatedWork W2788292821 @default.
- W3199280775 hasRelatedWork W2886673456 @default.
- W3199280775 hasRelatedWork W2912288872 @default.
- W3199280775 hasRelatedWork W3106036237 @default.
- W3199280775 hasRelatedWork W4285337767 @default.
- W3199280775 hasRelatedWork W564581980 @default.
- W3199280775 isParatext "false" @default.
- W3199280775 isRetracted "false" @default.
- W3199280775 magId "3199280775" @default.
- W3199280775 workType "article" @default.