Matches in SemOpenAlex for { <https://semopenalex.org/work/W4360955643> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W4360955643 endingPage "170809" @default.
- W4360955643 startingPage "170809" @default.
- W4360955643 abstract "In recent years, convolutional neural networks have become increasingly important in the field of machine learning, especially for computer vision. However, deep network models are difficult to deploy on hardware-constrained devices because of huge number of parameters, storage requirements and computational cost. This paper proposes a pruning feature maps method to delete redundant feature information in the deep network. Thus, it can simplify network structure at the same time reduce the computational complexity and speed up the operation. We first define a small chi-square supervised set. The feature maps of this set and the training set are extracted. Then, two variance matrices are constructed. The differences between the variance matrices are then used to establish chi-square distances. Through continuous experiments, the optimal position threshold is set, and the feature maps corresponding to the channels below the position threshold are pruned. Experiment performances show that the method can reduce network redundancy, storage space and network complexity totally up to 70%. This is accomplished without appreciably diminishing the network’s accuracy. In the worst case, the accuracy difference for images classification by using simplified network and network before simplification was less than 0.4%. The use of a small pre-trained network also speeds up the network training. Together, these improvements constitute an important step towards the effective implementation of CNNs on constrained devices." @default.
- W4360955643 created "2023-03-26" @default.
- W4360955643 creator A5002895137 @default.
- W4360955643 creator A5074082348 @default.
- W4360955643 creator A5090818211 @default.
- W4360955643 date "2023-06-01" @default.
- W4360955643 modified "2023-09-25" @default.
- W4360955643 title "Pruning feature maps for efficient convolutional neural networks" @default.
- W4360955643 cites W2104636679 @default.
- W4360955643 cites W2516041031 @default.
- W4360955643 cites W2788292821 @default.
- W4360955643 cites W2799390666 @default.
- W4360955643 cites W2899025077 @default.
- W4360955643 cites W2902716529 @default.
- W4360955643 cites W2963044089 @default.
- W4360955643 cites W2963145730 @default.
- W4360955643 cites W2963363373 @default.
- W4360955643 cites W2963681088 @default.
- W4360955643 cites W2981903848 @default.
- W4360955643 cites W3022592629 @default.
- W4360955643 cites W3040232122 @default.
- W4360955643 cites W3087618178 @default.
- W4360955643 cites W3167888193 @default.
- W4360955643 cites W3192416785 @default.
- W4360955643 cites W4200402413 @default.
- W4360955643 cites W4220902872 @default.
- W4360955643 cites W4225307160 @default.
- W4360955643 cites W4284967031 @default.
- W4360955643 cites W4302027380 @default.
- W4360955643 doi "https://doi.org/10.1016/j.ijleo.2023.170809" @default.
- W4360955643 hasPublicationYear "2023" @default.
- W4360955643 type Work @default.
- W4360955643 citedByCount "0" @default.
- W4360955643 crossrefType "journal-article" @default.
- W4360955643 hasAuthorship W4360955643A5002895137 @default.
- W4360955643 hasAuthorship W4360955643A5074082348 @default.
- W4360955643 hasAuthorship W4360955643A5090818211 @default.
- W4360955643 hasConcept C108010975 @default.
- W4360955643 hasConcept C111919701 @default.
- W4360955643 hasConcept C11413529 @default.
- W4360955643 hasConcept C124101348 @default.
- W4360955643 hasConcept C138885662 @default.
- W4360955643 hasConcept C152124472 @default.
- W4360955643 hasConcept C153180895 @default.
- W4360955643 hasConcept C154945302 @default.
- W4360955643 hasConcept C177264268 @default.
- W4360955643 hasConcept C179799912 @default.
- W4360955643 hasConcept C199360897 @default.
- W4360955643 hasConcept C2776401178 @default.
- W4360955643 hasConcept C41008148 @default.
- W4360955643 hasConcept C41895202 @default.
- W4360955643 hasConcept C50644808 @default.
- W4360955643 hasConcept C6557445 @default.
- W4360955643 hasConcept C81363708 @default.
- W4360955643 hasConcept C86803240 @default.
- W4360955643 hasConceptScore W4360955643C108010975 @default.
- W4360955643 hasConceptScore W4360955643C111919701 @default.
- W4360955643 hasConceptScore W4360955643C11413529 @default.
- W4360955643 hasConceptScore W4360955643C124101348 @default.
- W4360955643 hasConceptScore W4360955643C138885662 @default.
- W4360955643 hasConceptScore W4360955643C152124472 @default.
- W4360955643 hasConceptScore W4360955643C153180895 @default.
- W4360955643 hasConceptScore W4360955643C154945302 @default.
- W4360955643 hasConceptScore W4360955643C177264268 @default.
- W4360955643 hasConceptScore W4360955643C179799912 @default.
- W4360955643 hasConceptScore W4360955643C199360897 @default.
- W4360955643 hasConceptScore W4360955643C2776401178 @default.
- W4360955643 hasConceptScore W4360955643C41008148 @default.
- W4360955643 hasConceptScore W4360955643C41895202 @default.
- W4360955643 hasConceptScore W4360955643C50644808 @default.
- W4360955643 hasConceptScore W4360955643C6557445 @default.
- W4360955643 hasConceptScore W4360955643C81363708 @default.
- W4360955643 hasConceptScore W4360955643C86803240 @default.
- W4360955643 hasLocation W43609556431 @default.
- W4360955643 hasOpenAccess W4360955643 @default.
- W4360955643 hasPrimaryLocation W43609556431 @default.
- W4360955643 hasRelatedWork W2175746458 @default.
- W4360955643 hasRelatedWork W2732542196 @default.
- W4360955643 hasRelatedWork W2738221750 @default.
- W4360955643 hasRelatedWork W2760085659 @default.
- W4360955643 hasRelatedWork W2765989371 @default.
- W4360955643 hasRelatedWork W2767651786 @default.
- W4360955643 hasRelatedWork W2788292821 @default.
- W4360955643 hasRelatedWork W2883200793 @default.
- W4360955643 hasRelatedWork W3093612317 @default.
- W4360955643 hasRelatedWork W3094624326 @default.
- W4360955643 hasVolume "281" @default.
- W4360955643 isParatext "false" @default.
- W4360955643 isRetracted "false" @default.
- W4360955643 workType "article" @default.