Matches in SemOpenAlex for { <https://semopenalex.org/work/W2910343334> ?p ?o ?g. }
Showing items 1 to 83 of
83
with 100 items per page.
- W2910343334 abstract "Nowadays, the evolution of deep learning and cloud service significantly promotes neural network based mobile applications. Although intelligent and prolific, those applications still lack certain flexibility: For classification tasks, neural networks are generally trained online with vast classification targets to cover various utilization contexts. However, only partial classes are practically tested due to individual mobile user preference and application specificity. Thus the unneeded classes cause considerable computation and communication cost. In this work, we propose CAPTOR - a class-level reconfiguration framework for Convolutional Neural Networks (CNNs). By identifying the class activation preference of convolutional filters through feature interest visualization and gradient analysis, CAPTOR can effectively cluster and adaptively prune the filters associated with unneeded classes. Therefore, CAPTOR enables class-level CNN reconfiguration for network model compression and local deployment on mobile devices. Experiment shows that, CAPTOR can reduce computation load for VGG-16 by up to 40.5% and 37.9% energy consumption with ignored loss of accuracy. For AlexNet, CAPTOR also reduces computation load by up to 42.8% and 37.6% energy consumption with less than 3% loss in accuracy." @default.
- W2910343334 created "2019-01-25" @default.
- W2910343334 creator A5018832267 @default.
- W2910343334 creator A5060314107 @default.
- W2910343334 creator A5076768386 @default.
- W2910343334 creator A5082386342 @default.
- W2910343334 date "2019-01-21" @default.
- W2910343334 modified "2023-10-10" @default.
- W2910343334 title "CAPTOR" @default.
- W2910343334 cites W1849277567 @default.
- W2910343334 cites W1996901117 @default.
- W2910343334 cites W2525098248 @default.
- W2910343334 cites W2889185260 @default.
- W2910343334 doi "https://doi.org/10.1145/3287624.3287643" @default.
- W2910343334 hasPublicationYear "2019" @default.
- W2910343334 type Work @default.
- W2910343334 sameAs 2910343334 @default.
- W2910343334 citedByCount "6" @default.
- W2910343334 countsByYear W29103433342018 @default.
- W2910343334 countsByYear W29103433342020 @default.
- W2910343334 countsByYear W29103433342022 @default.
- W2910343334 crossrefType "proceedings-article" @default.
- W2910343334 hasAuthorship W2910343334A5018832267 @default.
- W2910343334 hasAuthorship W2910343334A5060314107 @default.
- W2910343334 hasAuthorship W2910343334A5076768386 @default.
- W2910343334 hasAuthorship W2910343334A5082386342 @default.
- W2910343334 hasBestOaLocation W29103433341 @default.
- W2910343334 hasConcept C111919701 @default.
- W2910343334 hasConcept C11413529 @default.
- W2910343334 hasConcept C115961682 @default.
- W2910343334 hasConcept C119599485 @default.
- W2910343334 hasConcept C119701452 @default.
- W2910343334 hasConcept C127413603 @default.
- W2910343334 hasConcept C138885662 @default.
- W2910343334 hasConcept C149635348 @default.
- W2910343334 hasConcept C153180895 @default.
- W2910343334 hasConcept C154945302 @default.
- W2910343334 hasConcept C186967261 @default.
- W2910343334 hasConcept C2776401178 @default.
- W2910343334 hasConcept C2780165032 @default.
- W2910343334 hasConcept C41008148 @default.
- W2910343334 hasConcept C41895202 @default.
- W2910343334 hasConcept C45374587 @default.
- W2910343334 hasConcept C52622490 @default.
- W2910343334 hasConcept C75294576 @default.
- W2910343334 hasConcept C81363708 @default.
- W2910343334 hasConceptScore W2910343334C111919701 @default.
- W2910343334 hasConceptScore W2910343334C11413529 @default.
- W2910343334 hasConceptScore W2910343334C115961682 @default.
- W2910343334 hasConceptScore W2910343334C119599485 @default.
- W2910343334 hasConceptScore W2910343334C119701452 @default.
- W2910343334 hasConceptScore W2910343334C127413603 @default.
- W2910343334 hasConceptScore W2910343334C138885662 @default.
- W2910343334 hasConceptScore W2910343334C149635348 @default.
- W2910343334 hasConceptScore W2910343334C153180895 @default.
- W2910343334 hasConceptScore W2910343334C154945302 @default.
- W2910343334 hasConceptScore W2910343334C186967261 @default.
- W2910343334 hasConceptScore W2910343334C2776401178 @default.
- W2910343334 hasConceptScore W2910343334C2780165032 @default.
- W2910343334 hasConceptScore W2910343334C41008148 @default.
- W2910343334 hasConceptScore W2910343334C41895202 @default.
- W2910343334 hasConceptScore W2910343334C45374587 @default.
- W2910343334 hasConceptScore W2910343334C52622490 @default.
- W2910343334 hasConceptScore W2910343334C75294576 @default.
- W2910343334 hasConceptScore W2910343334C81363708 @default.
- W2910343334 hasFunder F4320306076 @default.
- W2910343334 hasLocation W29103433341 @default.
- W2910343334 hasOpenAccess W2910343334 @default.
- W2910343334 hasPrimaryLocation W29103433341 @default.
- W2910343334 hasRelatedWork W2059299633 @default.
- W2910343334 hasRelatedWork W2406522397 @default.
- W2910343334 hasRelatedWork W2518599539 @default.
- W2910343334 hasRelatedWork W2726121760 @default.
- W2910343334 hasRelatedWork W2732542196 @default.
- W2910343334 hasRelatedWork W2760085659 @default.
- W2910343334 hasRelatedWork W2969680539 @default.
- W2910343334 hasRelatedWork W2977314777 @default.
- W2910343334 hasRelatedWork W2995914718 @default.
- W2910343334 hasRelatedWork W4307883119 @default.
- W2910343334 isParatext "false" @default.
- W2910343334 isRetracted "false" @default.
- W2910343334 magId "2910343334" @default.
- W2910343334 workType "article" @default.