Matches in SemOpenAlex for { <https://semopenalex.org/work/W4295369237> ?p ?o ?g. }
Showing items 1 to 69 of
69
with 100 items per page.
- W4295369237 abstract "Network pruning is aimed at imposing sparsity in a neural network architecture by increasing the portion of zero-valued weights for reducing its size regarding energy-efficiency consideration and increasing evaluation speed. In most of the conducted research efforts, the sparsity is enforced for network pruning without any attention to the internal network characteristics such as unbalanced outputs of the neurons or more specifically the distribution of the weights and outputs of the neurons. That may cause severe accuracy drop due to uncontrolled sparsity. In this work, we propose an attention mechanism that simultaneously controls the sparsity intensity and supervised network pruning by keeping important information bottlenecks of the network to be active. On CIFAR-10, the proposed method outperforms the best baseline method by 6% and reduced the accuracy drop by 2.6x at the same level of sparsity." @default.
- W4295369237 created "2022-09-13" @default.
- W4295369237 creator A5021852735 @default.
- W4295369237 creator A5033962469 @default.
- W4295369237 creator A5064784351 @default.
- W4295369237 creator A5076900687 @default.
- W4295369237 date "2018-02-12" @default.
- W4295369237 modified "2023-10-18" @default.
- W4295369237 title "Attention-Based Guided Structured Sparsity of Deep Neural Networks" @default.
- W4295369237 doi "https://doi.org/10.48550/arxiv.1802.09902" @default.
- W4295369237 hasPublicationYear "2018" @default.
- W4295369237 type Work @default.
- W4295369237 citedByCount "0" @default.
- W4295369237 crossrefType "posted-content" @default.
- W4295369237 hasAuthorship W4295369237A5021852735 @default.
- W4295369237 hasAuthorship W4295369237A5033962469 @default.
- W4295369237 hasAuthorship W4295369237A5064784351 @default.
- W4295369237 hasAuthorship W4295369237A5076900687 @default.
- W4295369237 hasBestOaLocation W42953692371 @default.
- W4295369237 hasConcept C105795698 @default.
- W4295369237 hasConcept C108010975 @default.
- W4295369237 hasConcept C111368507 @default.
- W4295369237 hasConcept C119857082 @default.
- W4295369237 hasConcept C12725497 @default.
- W4295369237 hasConcept C127313418 @default.
- W4295369237 hasConcept C153180895 @default.
- W4295369237 hasConcept C154945302 @default.
- W4295369237 hasConcept C186370098 @default.
- W4295369237 hasConcept C193415008 @default.
- W4295369237 hasConcept C2984842247 @default.
- W4295369237 hasConcept C31258907 @default.
- W4295369237 hasConcept C33923547 @default.
- W4295369237 hasConcept C41008148 @default.
- W4295369237 hasConcept C50644808 @default.
- W4295369237 hasConcept C6557445 @default.
- W4295369237 hasConcept C86803240 @default.
- W4295369237 hasConceptScore W4295369237C105795698 @default.
- W4295369237 hasConceptScore W4295369237C108010975 @default.
- W4295369237 hasConceptScore W4295369237C111368507 @default.
- W4295369237 hasConceptScore W4295369237C119857082 @default.
- W4295369237 hasConceptScore W4295369237C12725497 @default.
- W4295369237 hasConceptScore W4295369237C127313418 @default.
- W4295369237 hasConceptScore W4295369237C153180895 @default.
- W4295369237 hasConceptScore W4295369237C154945302 @default.
- W4295369237 hasConceptScore W4295369237C186370098 @default.
- W4295369237 hasConceptScore W4295369237C193415008 @default.
- W4295369237 hasConceptScore W4295369237C2984842247 @default.
- W4295369237 hasConceptScore W4295369237C31258907 @default.
- W4295369237 hasConceptScore W4295369237C33923547 @default.
- W4295369237 hasConceptScore W4295369237C41008148 @default.
- W4295369237 hasConceptScore W4295369237C50644808 @default.
- W4295369237 hasConceptScore W4295369237C6557445 @default.
- W4295369237 hasConceptScore W4295369237C86803240 @default.
- W4295369237 hasLocation W42953692371 @default.
- W4295369237 hasOpenAccess W4295369237 @default.
- W4295369237 hasPrimaryLocation W42953692371 @default.
- W4295369237 hasRelatedWork W1490470283 @default.
- W4295369237 hasRelatedWork W2950066684 @default.
- W4295369237 hasRelatedWork W3199608561 @default.
- W4295369237 hasRelatedWork W3203168320 @default.
- W4295369237 hasRelatedWork W4206024927 @default.
- W4295369237 hasRelatedWork W4288853838 @default.
- W4295369237 hasRelatedWork W4298388782 @default.
- W4295369237 hasRelatedWork W4322008322 @default.
- W4295369237 hasRelatedWork W4367297978 @default.
- W4295369237 hasRelatedWork W1629725936 @default.
- W4295369237 isParatext "false" @default.
- W4295369237 isRetracted "false" @default.
- W4295369237 workType "article" @default.