FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4322773516> ?p ?o ?g. }

• W4322773516 abstract "Filter pruning is widely used for inference acceleration and compatibility with off-the-shelf hardware devices. Some filter pruning methods have proposed various criteria to approximate the importance of filters, and then sort the filters globally or locally to prune the redundant parameters. However, the current criterion-based methods have problems: (1) parameters with smaller criterion values for extracting edge features are easily ignored, and (2) there is a strong correlation between different criteria, resulting in similar pruning structures. In this article, we propose a novel simple but effective pruning method based on filter similarity, which is used to evaluate the similarity between filters instead of the importance of a single filter. The proposed method first calculates the similarity of the filters pairwise in one convolutional layer and then obtains the similarity distribution. Finally, the filters with high similarity to others are deleted from the distribution or set to zero. In addition, the proposed algorithm does not need to specify the pruning rate for each layer, and only needs to set the desired FLOPs or parameter reduction to obtain the final compression model. We also provide iterative pruning strategies for hard pruning and soft pruning to satisfy the tradeoff requirements of accuracy and memory in different scenarios. Extensive experiments on various representative benchmark datasets across different network architectures demonstrate the effectiveness of our proposed method. For example, on CIFAR10, the proposed algorithm achieves 61.1% FLOPs reduction by removing 58.3% of the parameters, with no loss in Top-1 accuracy on ResNet-56; and reduces 53.05% FLOPs on ResNet-50 with only 0.29% Top-1 accuracy degradation on ILSVRC-2012." @default.
• W4322773516 created "2023-03-03" @default.
• W4322773516 creator A5005866333 @default.
• W4322773516 creator A5043075442 @default.
• W4322773516 creator A5070221266 @default.
• W4322773516 date "2023-03-02" @default.
• W4322773516 modified "2023-09-26" @default.
• W4322773516 title "Model pruning based on filter similarity for edge device deployment" @default.
• W4322773516 cites W2117539524 @default.
• W4322773516 cites W2143612262 @default.
• W4322773516 cites W2194775991 @default.
• W4322773516 cites W2495425901 @default.
• W4322773516 cites W2604998962 @default.
• W4322773516 cites W2618530766 @default.
• W4322773516 cites W2791026767 @default.
• W4322773516 cites W2808168148 @default.
• W4322773516 cites W2883111419 @default.
• W4322773516 cites W2886851211 @default.
• W4322773516 cites W2928560789 @default.
• W4322773516 cites W2962851801 @default.
• W4322773516 cites W2962965870 @default.
• W4322773516 cites W2963125010 @default.
• W4322773516 cites W2963145730 @default.
• W4322773516 cites W2963363373 @default.
• W4322773516 cites W2964001144 @default.
• W4322773516 cites W2964233199 @default.
• W4322773516 cites W2964502788 @default.
• W4322773516 cites W2970958999 @default.
• W4322773516 cites W3018228401 @default.
• W4322773516 cites W3034368386 @default.
• W4322773516 cites W3034513523 @default.
• W4322773516 cites W3037246750 @default.
• W4322773516 cites W3038173901 @default.
• W4322773516 cites W3043678861 @default.
• W4322773516 cites W3162666703 @default.
• W4322773516 cites W3181161645 @default.
• W4322773516 cites W4226350879 @default.
• W4322773516 cites W4283815570 @default.
• W4322773516 cites W4297775537 @default.
• W4322773516 cites W639708223 @default.
• W4322773516 doi "https://doi.org/10.3389/fnbot.2023.1132679" @default.
• W4322773516 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/36937554" @default.
• W4322773516 hasPublicationYear "2023" @default.
• W4322773516 type Work @default.
• W4322773516 citedByCount "0" @default.
• W4322773516 crossrefType "journal-article" @default.
• W4322773516 hasAuthorship W4322773516A5005866333 @default.
• W4322773516 hasAuthorship W4322773516A5043075442 @default.
• W4322773516 hasAuthorship W4322773516A5070221266 @default.
• W4322773516 hasBestOaLocation W43227735161 @default.
• W4322773516 hasConcept C106131492 @default.
• W4322773516 hasConcept C108010975 @default.
• W4322773516 hasConcept C111335779 @default.
• W4322773516 hasConcept C11413529 @default.
• W4322773516 hasConcept C13280743 @default.
• W4322773516 hasConcept C153180895 @default.
• W4322773516 hasConcept C154945302 @default.
• W4322773516 hasConcept C185798385 @default.
• W4322773516 hasConcept C205649164 @default.
• W4322773516 hasConcept C2524010 @default.
• W4322773516 hasConcept C31972630 @default.
• W4322773516 hasConcept C33923547 @default.
• W4322773516 hasConcept C41008148 @default.
• W4322773516 hasConcept C6557445 @default.
• W4322773516 hasConcept C86803240 @default.
• W4322773516 hasConceptScore W4322773516C106131492 @default.
• W4322773516 hasConceptScore W4322773516C108010975 @default.
• W4322773516 hasConceptScore W4322773516C111335779 @default.
• W4322773516 hasConceptScore W4322773516C11413529 @default.
• W4322773516 hasConceptScore W4322773516C13280743 @default.
• W4322773516 hasConceptScore W4322773516C153180895 @default.
• W4322773516 hasConceptScore W4322773516C154945302 @default.
• W4322773516 hasConceptScore W4322773516C185798385 @default.
• W4322773516 hasConceptScore W4322773516C205649164 @default.
• W4322773516 hasConceptScore W4322773516C2524010 @default.
• W4322773516 hasConceptScore W4322773516C31972630 @default.
• W4322773516 hasConceptScore W4322773516C33923547 @default.
• W4322773516 hasConceptScore W4322773516C41008148 @default.
• W4322773516 hasConceptScore W4322773516C6557445 @default.
• W4322773516 hasConceptScore W4322773516C86803240 @default.
• W4322773516 hasFunder F4320321001 @default.
• W4322773516 hasFunder F4320335777 @default.
• W4322773516 hasLocation W43227735161 @default.
• W4322773516 hasLocation W43227735162 @default.
• W4322773516 hasLocation W43227735163 @default.
• W4322773516 hasOpenAccess W4322773516 @default.
• W4322773516 hasPrimaryLocation W43227735161 @default.
• W4322773516 hasRelatedWork W108084911 @default.
• W4322773516 hasRelatedWork W112744582 @default.
• W4322773516 hasRelatedWork W1485630101 @default.
• W4322773516 hasRelatedWork W2468451894 @default.
• W4322773516 hasRelatedWork W2498017833 @default.
• W4322773516 hasRelatedWork W2899084033 @default.
• W4322773516 hasRelatedWork W3047144510 @default.
• W4322773516 hasRelatedWork W4306294911 @default.
• W4322773516 hasRelatedWork W4367190475 @default.
• W4322773516 hasRelatedWork W4372260272 @default.
• W4322773516 hasVolume "17" @default.
• W4322773516 isParatext "false" @default.
• W4322773516 isRetracted "false" @default.

• next