FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2769044225> ?p ?o ?g. }

• W2769044225 abstract "To reduce the significant redundancy in deep Convolutional Neural Networks (CNNs), most existing methods prune neurons by only considering statistics of an individual layer or two consecutive layers (e.g., prune one layer to minimize the reconstruction error of the next layer), ignoring the effect of error propagation in deep networks. In contrast, we argue that it is essential to prune neurons in the entire neuron network jointly based on a unified goal: minimizing the reconstruction error of important responses in the final response layer (FRL), which is the second-to-last layer before classification, for a pruned network to retrain its predictive power. Specifically, we apply feature ranking techniques to measure the importance of each neuron in the FRL, and formulate network pruning as a binary integer optimization problem and derive a closed-form solution to it for pruning neurons in earlier layers. Based on our theoretical analysis, we propose the Neuron Importance Score Propagation (NISP) algorithm to propagate the importance scores of final responses to every neuron in the network. The CNN is pruned by removing neurons with least importance, and then fine-tuned to retain its predictive power. NISP is evaluated on several datasets with multiple CNN models and demonstrated to achieve significant acceleration and compression with negligible accuracy loss." @default.
• W2769044225 created "2017-12-04" @default.
• W2769044225 creator A5026664462 @default.
• W2769044225 creator A5038267072 @default.
• W2769044225 creator A5046330465 @default.
• W2769044225 creator A5055264952 @default.
• W2769044225 creator A5059605240 @default.
• W2769044225 creator A5080417608 @default.
• W2769044225 creator A5082572337 @default.
• W2769044225 creator A5086390754 @default.
• W2769044225 creator A5089030767 @default.
• W2769044225 date "2017-11-15" @default.
• W2769044225 modified "2023-09-25" @default.
• W2769044225 title "NISP: Pruning Networks using Neuron Importance Score Propagation" @default.
• W2769044225 cites W1532826273 @default.
• W2769044225 cites W1821462560 @default.
• W2769044225 cites W1825672851 @default.
• W2769044225 cites W1902041153 @default.
• W2769044225 cites W1935978687 @default.
• W2769044225 cites W2097117768 @default.
• W2769044225 cites W2108598243 @default.
• W2769044225 cites W2114766824 @default.
• W2769044225 cites W2125389748 @default.
• W2769044225 cites W2152332944 @default.
• W2769044225 cites W2167215970 @default.
• W2769044225 cites W2181028625 @default.
• W2769044225 cites W2194775991 @default.
• W2769044225 cites W2310919327 @default.
• W2769044225 cites W2331501241 @default.
• W2769044225 cites W2513419314 @default.
• W2769044225 cites W2519163251 @default.
• W2769044225 cites W2558275547 @default.
• W2769044225 cites W2740693737 @default.
• W2769044225 cites W2768169767 @default.
• W2769044225 cites W2774245448 @default.
• W2769044225 cites W2798357659 @default.
• W2769044225 cites W2949560654 @default.
• W2769044225 cites W2950094539 @default.
• W2769044225 cites W2950967261 @default.
• W2769044225 cites W2952432176 @default.
• W2769044225 cites W2952899695 @default.
• W2769044225 cites W2953106684 @default.
• W2769044225 cites W2962944050 @default.
• W2769044225 cites W2962988160 @default.
• W2769044225 cites W2963422144 @default.
• W2769044225 cites W2963897760 @default.
• W2769044225 cites W2964050021 @default.
• W2769044225 cites W2964233199 @default.
• W2769044225 cites W2964299589 @default.
• W2769044225 cites W3106250896 @default.
• W2769044225 cites W3118608800 @default.
• W2769044225 cites W992687842 @default.
• W2769044225 doi "https://doi.org/10.48550/arxiv.1711.05908" @default.
• W2769044225 hasPublicationYear "2017" @default.
• W2769044225 type Work @default.
• W2769044225 sameAs 2769044225 @default.
• W2769044225 citedByCount "9" @default.
• W2769044225 countsByYear W27690442252017 @default.
• W2769044225 countsByYear W27690442252018 @default.
• W2769044225 countsByYear W27690442252019 @default.
• W2769044225 countsByYear W27690442252020 @default.
• W2769044225 countsByYear W27690442252021 @default.
• W2769044225 crossrefType "posted-content" @default.
• W2769044225 hasAuthorship W2769044225A5026664462 @default.
• W2769044225 hasAuthorship W2769044225A5038267072 @default.
• W2769044225 hasAuthorship W2769044225A5046330465 @default.
• W2769044225 hasAuthorship W2769044225A5055264952 @default.
• W2769044225 hasAuthorship W2769044225A5059605240 @default.
• W2769044225 hasAuthorship W2769044225A5080417608 @default.
• W2769044225 hasAuthorship W2769044225A5082572337 @default.
• W2769044225 hasAuthorship W2769044225A5086390754 @default.
• W2769044225 hasAuthorship W2769044225A5089030767 @default.
• W2769044225 hasBestOaLocation W27690442251 @default.
• W2769044225 hasConcept C108010975 @default.
• W2769044225 hasConcept C111919701 @default.
• W2769044225 hasConcept C11413529 @default.
• W2769044225 hasConcept C138885662 @default.
• W2769044225 hasConcept C152124472 @default.
• W2769044225 hasConcept C153180895 @default.
• W2769044225 hasConcept C154945302 @default.
• W2769044225 hasConcept C178790620 @default.
• W2769044225 hasConcept C185592680 @default.
• W2769044225 hasConcept C189430467 @default.
• W2769044225 hasConcept C2776401178 @default.
• W2769044225 hasConcept C2779227376 @default.
• W2769044225 hasConcept C33923547 @default.
• W2769044225 hasConcept C41008148 @default.
• W2769044225 hasConcept C41895202 @default.
• W2769044225 hasConcept C48372109 @default.
• W2769044225 hasConcept C50644808 @default.
• W2769044225 hasConcept C6557445 @default.
• W2769044225 hasConcept C81363708 @default.
• W2769044225 hasConcept C86803240 @default.
• W2769044225 hasConcept C94375191 @default.
• W2769044225 hasConceptScore W2769044225C108010975 @default.
• W2769044225 hasConceptScore W2769044225C111919701 @default.
• W2769044225 hasConceptScore W2769044225C11413529 @default.
• W2769044225 hasConceptScore W2769044225C138885662 @default.
• W2769044225 hasConceptScore W2769044225C152124472 @default.
• W2769044225 hasConceptScore W2769044225C153180895 @default.

• next