FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387675809> ?p ?o ?g. }

• W4387675809 endingPage "122180" @default.
• W4387675809 startingPage "122180" @default.
• W4387675809 abstract "During recent years, deep convolutional neural networks (CNNs) have become increasingly popular in sensor-based human activity recognition (HAR), due to powerful feature extraction capacity. However, the required high computational cost has hindered their practical deployment in real-world applications. To address this issue, this paper presents a novel approach named ABCSearch to prune channels through an artificial bee colony (ABC) algorithm, which is a meta-heuristic optimization algorithm inspired by the foraging behavior of honeybees. Unlike traditional pruning strategies that prioritize channel importance, the proposed method seeks to identify the optimal number of channels at each layer, i.e., the optimal pruning structure. To lessen human interference, we transform the search problem of the optimal pruning structure into a simple optimization problem and solve it automatically through the ABC algorithm, which can significantly decrease both model size and computational cost without compromising performance. Specifically, we validate the effectiveness of ABCSearch for a standard five-layer CNN backbone architecture, which achieves 97.05%, 98.45%, 91.71%, 78.89% classification accuracies on four HAR benchmarks, i.e., UCI-HAR, WISDM, PAMAP2, UniMiB-SHAR. Our approach can yield 87.44%, 89.41%, 89.06%, 89.59% FLOPs reductions, and 89.54%, 88.70%, 88.38%, 90.45% less memory consumptions, as well as 4.4×, 5.0×, 5.8×, 6.1× practical speedups on real hardware devices, while maintaining comparable accuracy." @default.
• W4387675809 created "2023-10-17" @default.
• W4387675809 creator A5005207362 @default.
• W4387675809 creator A5009342809 @default.
• W4387675809 creator A5023925791 @default.
• W4387675809 creator A5038581639 @default.
• W4387675809 creator A5076127071 @default.
• W4387675809 creator A5090593667 @default.
• W4387675809 date "2023-10-01" @default.
• W4387675809 modified "2023-10-17" @default.
• W4387675809 title "An automatic network structure search via channel pruning for accelerating human activity inference on mobile devices" @default.
• W4387675809 cites W2017634428 @default.
• W4387675809 cites W2023302299 @default.
• W4387675809 cites W2038950936 @default.
• W4387675809 cites W2121269968 @default.
• W4387675809 cites W2169064301 @default.
• W4387675809 cites W2342792048 @default.
• W4387675809 cites W2555209581 @default.
• W4387675809 cites W2563686712 @default.
• W4387675809 cites W2618530766 @default.
• W4387675809 cites W2759690896 @default.
• W4387675809 cites W2795342689 @default.
• W4387675809 cites W2972317743 @default.
• W4387675809 cites W3034368386 @default.
• W4387675809 cites W3091587795 @default.
• W4387675809 cites W3100321043 @default.
• W4387675809 cites W3160306726 @default.
• W4387675809 cites W3211522713 @default.
• W4387675809 cites W3215341388 @default.
• W4387675809 cites W4200317051 @default.
• W4387675809 cites W4220659807 @default.
• W4387675809 cites W4317495236 @default.
• W4387675809 cites W4378189401 @default.
• W4387675809 cites W4381485879 @default.
• W4387675809 doi "https://doi.org/10.1016/j.eswa.2023.122180" @default.
• W4387675809 hasPublicationYear "2023" @default.
• W4387675809 type Work @default.
• W4387675809 citedByCount "0" @default.
• W4387675809 crossrefType "journal-article" @default.
• W4387675809 hasAuthorship W4387675809A5005207362 @default.
• W4387675809 hasAuthorship W4387675809A5009342809 @default.
• W4387675809 hasAuthorship W4387675809A5023925791 @default.
• W4387675809 hasAuthorship W4387675809A5038581639 @default.
• W4387675809 hasAuthorship W4387675809A5076127071 @default.
• W4387675809 hasAuthorship W4387675809A5090593667 @default.
• W4387675809 hasConcept C108010975 @default.
• W4387675809 hasConcept C108583219 @default.
• W4387675809 hasConcept C111472728 @default.
• W4387675809 hasConcept C119857082 @default.
• W4387675809 hasConcept C127162648 @default.
• W4387675809 hasConcept C138885662 @default.
• W4387675809 hasConcept C153180895 @default.
• W4387675809 hasConcept C154945302 @default.
• W4387675809 hasConcept C173608175 @default.
• W4387675809 hasConcept C173801870 @default.
• W4387675809 hasConcept C178790620 @default.
• W4387675809 hasConcept C185592680 @default.
• W4387675809 hasConcept C2776214188 @default.
• W4387675809 hasConcept C2776401178 @default.
• W4387675809 hasConcept C2779227376 @default.
• W4387675809 hasConcept C2780586882 @default.
• W4387675809 hasConcept C31258907 @default.
• W4387675809 hasConcept C3826847 @default.
• W4387675809 hasConcept C41008148 @default.
• W4387675809 hasConcept C41895202 @default.
• W4387675809 hasConcept C6557445 @default.
• W4387675809 hasConcept C81363708 @default.
• W4387675809 hasConcept C86803240 @default.
• W4387675809 hasConceptScore W4387675809C108010975 @default.
• W4387675809 hasConceptScore W4387675809C108583219 @default.
• W4387675809 hasConceptScore W4387675809C111472728 @default.
• W4387675809 hasConceptScore W4387675809C119857082 @default.
• W4387675809 hasConceptScore W4387675809C127162648 @default.
• W4387675809 hasConceptScore W4387675809C138885662 @default.
• W4387675809 hasConceptScore W4387675809C153180895 @default.
• W4387675809 hasConceptScore W4387675809C154945302 @default.
• W4387675809 hasConceptScore W4387675809C173608175 @default.
• W4387675809 hasConceptScore W4387675809C173801870 @default.
• W4387675809 hasConceptScore W4387675809C178790620 @default.
• W4387675809 hasConceptScore W4387675809C185592680 @default.
• W4387675809 hasConceptScore W4387675809C2776214188 @default.
• W4387675809 hasConceptScore W4387675809C2776401178 @default.
• W4387675809 hasConceptScore W4387675809C2779227376 @default.
• W4387675809 hasConceptScore W4387675809C2780586882 @default.
• W4387675809 hasConceptScore W4387675809C31258907 @default.
• W4387675809 hasConceptScore W4387675809C3826847 @default.
• W4387675809 hasConceptScore W4387675809C41008148 @default.
• W4387675809 hasConceptScore W4387675809C41895202 @default.
• W4387675809 hasConceptScore W4387675809C6557445 @default.
• W4387675809 hasConceptScore W4387675809C81363708 @default.
• W4387675809 hasConceptScore W4387675809C86803240 @default.
• W4387675809 hasLocation W43876758091 @default.
• W4387675809 hasOpenAccess W4387675809 @default.
• W4387675809 hasPrimaryLocation W43876758091 @default.
• W4387675809 hasRelatedWork W2971502891 @default.
• W4387675809 hasRelatedWork W2986126107 @default.
• W4387675809 hasRelatedWork W2996519767 @default.
• W4387675809 hasRelatedWork W3092292339 @default.

• next