Matches in SemOpenAlex for { <https://semopenalex.org/work/W4362676543> ?p ?o ?g. }
Showing items 1 to 91 of
91
with 100 items per page.
- W4362676543 endingPage "5791" @default.
- W4362676543 startingPage "5779" @default.
- W4362676543 abstract "Abstract Although the classification method based on the deep neural network has achieved excellent results in classification tasks, it is difficult to apply to real-time scenarios because of high memory footprints and prohibitive inference times. Compared to unstructured pruning, structured pruning techniques can reduce the computation cost of the model runtime more effectively, but inevitably reduces the precision of the model. Traditional methods use fine tuning to restore model damage performance. However, there is still a large gap between the pruned model and the original one. In this paper, we use progressive multi-level distillation learning to compensate for the loss caused by pruning. Pre-pruning and post-pruning networks serve as the teacher and student networks. The proposed approach utilizes the complementary properties of structured pruning and knowledge distillation, which allows the pruned network to learn the intermediate and output representations of the teacher network, thus reducing the influence of the model subject to pruning. Experiments demonstrate that our approach performs better on CIFAR-10, CIFAR-100, and Tiny-ImageNet datasets with different pruning rates. For instance, GoogLeNet can achieve near lossless pruning on the CIFAR-10 dataset with 60% pruning. Moreover, this paper also proves that using the proposed distillation learning method during the pruning process achieves more significant performance gains than after completing the pruning." @default.
- W4362676543 created "2023-04-07" @default.
- W4362676543 creator A5003835661 @default.
- W4362676543 creator A5013235835 @default.
- W4362676543 creator A5018073672 @default.
- W4362676543 creator A5019041035 @default.
- W4362676543 creator A5029239459 @default.
- W4362676543 creator A5044016780 @default.
- W4362676543 creator A5056623744 @default.
- W4362676543 creator A5063565261 @default.
- W4362676543 creator A5064144631 @default.
- W4362676543 creator A5089758953 @default.
- W4362676543 date "2023-04-07" @default.
- W4362676543 modified "2023-10-17" @default.
- W4362676543 title "Progressive multi-level distillation learning for pruning network" @default.
- W4362676543 cites W2112796928 @default.
- W4362676543 cites W2285660444 @default.
- W4362676543 cites W2884001105 @default.
- W4362676543 cites W2964137095 @default.
- W4362676543 cites W3004543888 @default.
- W4362676543 cites W3034368386 @default.
- W4362676543 cites W3042620538 @default.
- W4362676543 cites W3171155106 @default.
- W4362676543 cites W3196438592 @default.
- W4362676543 cites W3201277138 @default.
- W4362676543 doi "https://doi.org/10.1007/s40747-023-01036-0" @default.
- W4362676543 hasPublicationYear "2023" @default.
- W4362676543 type Work @default.
- W4362676543 citedByCount "1" @default.
- W4362676543 countsByYear W43626765432023 @default.
- W4362676543 crossrefType "journal-article" @default.
- W4362676543 hasAuthorship W4362676543A5003835661 @default.
- W4362676543 hasAuthorship W4362676543A5013235835 @default.
- W4362676543 hasAuthorship W4362676543A5018073672 @default.
- W4362676543 hasAuthorship W4362676543A5019041035 @default.
- W4362676543 hasAuthorship W4362676543A5029239459 @default.
- W4362676543 hasAuthorship W4362676543A5044016780 @default.
- W4362676543 hasAuthorship W4362676543A5056623744 @default.
- W4362676543 hasAuthorship W4362676543A5063565261 @default.
- W4362676543 hasAuthorship W4362676543A5064144631 @default.
- W4362676543 hasAuthorship W4362676543A5089758953 @default.
- W4362676543 hasBestOaLocation W43626765431 @default.
- W4362676543 hasConcept C108010975 @default.
- W4362676543 hasConcept C108583219 @default.
- W4362676543 hasConcept C111919701 @default.
- W4362676543 hasConcept C119857082 @default.
- W4362676543 hasConcept C154945302 @default.
- W4362676543 hasConcept C178790620 @default.
- W4362676543 hasConcept C185592680 @default.
- W4362676543 hasConcept C204030448 @default.
- W4362676543 hasConcept C2776214188 @default.
- W4362676543 hasConcept C41008148 @default.
- W4362676543 hasConcept C50644808 @default.
- W4362676543 hasConcept C6557445 @default.
- W4362676543 hasConcept C86803240 @default.
- W4362676543 hasConcept C98045186 @default.
- W4362676543 hasConceptScore W4362676543C108010975 @default.
- W4362676543 hasConceptScore W4362676543C108583219 @default.
- W4362676543 hasConceptScore W4362676543C111919701 @default.
- W4362676543 hasConceptScore W4362676543C119857082 @default.
- W4362676543 hasConceptScore W4362676543C154945302 @default.
- W4362676543 hasConceptScore W4362676543C178790620 @default.
- W4362676543 hasConceptScore W4362676543C185592680 @default.
- W4362676543 hasConceptScore W4362676543C204030448 @default.
- W4362676543 hasConceptScore W4362676543C2776214188 @default.
- W4362676543 hasConceptScore W4362676543C41008148 @default.
- W4362676543 hasConceptScore W4362676543C50644808 @default.
- W4362676543 hasConceptScore W4362676543C6557445 @default.
- W4362676543 hasConceptScore W4362676543C86803240 @default.
- W4362676543 hasConceptScore W4362676543C98045186 @default.
- W4362676543 hasFunder F4320336578 @default.
- W4362676543 hasIssue "5" @default.
- W4362676543 hasLocation W43626765431 @default.
- W4362676543 hasOpenAccess W4362676543 @default.
- W4362676543 hasPrimaryLocation W43626765431 @default.
- W4362676543 hasRelatedWork W2795261237 @default.
- W4362676543 hasRelatedWork W3014300295 @default.
- W4362676543 hasRelatedWork W3164822677 @default.
- W4362676543 hasRelatedWork W4223943233 @default.
- W4362676543 hasRelatedWork W4225161397 @default.
- W4362676543 hasRelatedWork W4312200629 @default.
- W4362676543 hasRelatedWork W4360585206 @default.
- W4362676543 hasRelatedWork W4364306694 @default.
- W4362676543 hasRelatedWork W4380075502 @default.
- W4362676543 hasRelatedWork W4380086463 @default.
- W4362676543 hasVolume "9" @default.
- W4362676543 isParatext "false" @default.
- W4362676543 isRetracted "false" @default.
- W4362676543 workType "article" @default.