Matches in SemOpenAlex for { <https://semopenalex.org/work/W3024605841> ?p ?o ?g. }
Showing items 1 to 99 of
99
with 100 items per page.
- W3024605841 endingPage "90930" @default.
- W3024605841 startingPage "90924" @default.
- W3024605841 abstract "Due to its over-parameterized design, the deep convolutional neural networks lead to a huge amount of parameters and high computational cost, making it difficult to deploy on some devices with limited computational resources in reality. In this paper, we propose a method of filter pruning without damaging networks capacity to accelerate and compress deep convolutional neural networks. Differently from some existing filter pruning methods, we pay more attention to the damage by filter pruning to model capacity. In order to restore the original model capacity, we generate new feature maps on the basis of the remaining feature maps with lighter structure after pruning the redundant filters that are similar with the others. Experimental results on CIFAR10 and CIFAR100 benchmarks demonstrate the effectiveness of our method. Especially, our method reduces more than 49% FLOPs for VGGNet-16 on CIFAR10 with only 0.07% relative accuracy drop. The relative accuracy has even been increased by 0.13% with reducing more than 24% FLOPs. Moreover, our method accelerates ResNet-110 on CIFAR10 by 22.1% with 0.41% accuracy improvement, which exceeds the previous methods." @default.
- W3024605841 created "2020-05-21" @default.
- W3024605841 creator A5020016757 @default.
- W3024605841 creator A5027874520 @default.
- W3024605841 creator A5049447178 @default.
- W3024605841 creator A5060530625 @default.
- W3024605841 date "2020-01-01" @default.
- W3024605841 modified "2023-09-27" @default.
- W3024605841 title "Filter Pruning Without Damaging Networks Capacity" @default.
- W3024605841 cites W1536680647 @default.
- W3024605841 cites W1745334888 @default.
- W3024605841 cites W1996901117 @default.
- W3024605841 cites W2194775991 @default.
- W3024605841 cites W2302255633 @default.
- W3024605841 cites W2798603777 @default.
- W3024605841 cites W2799197246 @default.
- W3024605841 cites W2808168148 @default.
- W3024605841 cites W2890788829 @default.
- W3024605841 cites W2928560789 @default.
- W3024605841 cites W2962851801 @default.
- W3024605841 cites W2963122961 @default.
- W3024605841 cites W2963363373 @default.
- W3024605841 cites W2964233199 @default.
- W3024605841 cites W2965862774 @default.
- W3024605841 cites W2995104385 @default.
- W3024605841 doi "https://doi.org/10.1109/access.2020.2993932" @default.
- W3024605841 hasPublicationYear "2020" @default.
- W3024605841 type Work @default.
- W3024605841 sameAs 3024605841 @default.
- W3024605841 citedByCount "12" @default.
- W3024605841 countsByYear W30246058412020 @default.
- W3024605841 countsByYear W30246058412021 @default.
- W3024605841 countsByYear W30246058412022 @default.
- W3024605841 countsByYear W30246058412023 @default.
- W3024605841 crossrefType "journal-article" @default.
- W3024605841 hasAuthorship W3024605841A5020016757 @default.
- W3024605841 hasAuthorship W3024605841A5027874520 @default.
- W3024605841 hasAuthorship W3024605841A5049447178 @default.
- W3024605841 hasAuthorship W3024605841A5060530625 @default.
- W3024605841 hasBestOaLocation W30246058411 @default.
- W3024605841 hasConcept C106131492 @default.
- W3024605841 hasConcept C108010975 @default.
- W3024605841 hasConcept C108583219 @default.
- W3024605841 hasConcept C11413529 @default.
- W3024605841 hasConcept C119857082 @default.
- W3024605841 hasConcept C138885662 @default.
- W3024605841 hasConcept C153180895 @default.
- W3024605841 hasConcept C154945302 @default.
- W3024605841 hasConcept C173608175 @default.
- W3024605841 hasConcept C2776401178 @default.
- W3024605841 hasConcept C31972630 @default.
- W3024605841 hasConcept C3826847 @default.
- W3024605841 hasConcept C41008148 @default.
- W3024605841 hasConcept C41895202 @default.
- W3024605841 hasConcept C45347329 @default.
- W3024605841 hasConcept C50644808 @default.
- W3024605841 hasConcept C6557445 @default.
- W3024605841 hasConcept C81363708 @default.
- W3024605841 hasConcept C86803240 @default.
- W3024605841 hasConceptScore W3024605841C106131492 @default.
- W3024605841 hasConceptScore W3024605841C108010975 @default.
- W3024605841 hasConceptScore W3024605841C108583219 @default.
- W3024605841 hasConceptScore W3024605841C11413529 @default.
- W3024605841 hasConceptScore W3024605841C119857082 @default.
- W3024605841 hasConceptScore W3024605841C138885662 @default.
- W3024605841 hasConceptScore W3024605841C153180895 @default.
- W3024605841 hasConceptScore W3024605841C154945302 @default.
- W3024605841 hasConceptScore W3024605841C173608175 @default.
- W3024605841 hasConceptScore W3024605841C2776401178 @default.
- W3024605841 hasConceptScore W3024605841C31972630 @default.
- W3024605841 hasConceptScore W3024605841C3826847 @default.
- W3024605841 hasConceptScore W3024605841C41008148 @default.
- W3024605841 hasConceptScore W3024605841C41895202 @default.
- W3024605841 hasConceptScore W3024605841C45347329 @default.
- W3024605841 hasConceptScore W3024605841C50644808 @default.
- W3024605841 hasConceptScore W3024605841C6557445 @default.
- W3024605841 hasConceptScore W3024605841C81363708 @default.
- W3024605841 hasConceptScore W3024605841C86803240 @default.
- W3024605841 hasLocation W30246058411 @default.
- W3024605841 hasLocation W30246058412 @default.
- W3024605841 hasOpenAccess W3024605841 @default.
- W3024605841 hasPrimaryLocation W30246058411 @default.
- W3024605841 hasRelatedWork W2295021132 @default.
- W3024605841 hasRelatedWork W2731899572 @default.
- W3024605841 hasRelatedWork W2898360562 @default.
- W3024605841 hasRelatedWork W3047530476 @default.
- W3024605841 hasRelatedWork W3133861977 @default.
- W3024605841 hasRelatedWork W3183066147 @default.
- W3024605841 hasRelatedWork W4200173597 @default.
- W3024605841 hasRelatedWork W4312417841 @default.
- W3024605841 hasRelatedWork W4321369474 @default.
- W3024605841 hasRelatedWork W4386399705 @default.
- W3024605841 hasVolume "8" @default.
- W3024605841 isParatext "false" @default.
- W3024605841 isRetracted "false" @default.
- W3024605841 magId "3024605841" @default.
- W3024605841 workType "article" @default.