Matches in SemOpenAlex for { <https://semopenalex.org/work/W2897623300> ?p ?o ?g. }
Showing items 1 to 92 of
92
with 100 items per page.
- W2897623300 abstract "Deep convolutional neural networks (ConvNets) have achieved the state-of-the-art performance on many real-world applications. However, significant computation and storage demands are required by ConvNets. Spiking neural networks (SNNs), with sparsely activated neurons and event-driven computations, show great potential to take advantage of the ultra- low power spike-based hardware architectures. Yet, training SNN with similar accuracy as ConvNets is difficult. Recent researchers have demonstrated the work of converting ConvNets to SNNs (CNN-SNN conversion) with similar accuracy. However, the energy-efficiency of the converted SNNs is impaired by the increased classification latency. In this paper, we focus on optimizing the classification latency of the converted SNNs. First, we propose a restricted output training method to normalize the converted weights dynamically in the CNN-SNN training phase. Second, false spikes are identified and the false spike inhibition theory is derived to speedup the convergence of the classification process. Third, we propose a temporal max pooling method to approximate the max pooling operation in ConvNets without accuracy loss. The evaluation shows that the converted SNNs converge in about 30 time-steps and achieve the best classification accuracy of 94% on CIFAR -10 dataset." @default.
- W2897623300 created "2018-10-26" @default.
- W2897623300 creator A5000050947 @default.
- W2897623300 creator A5002976537 @default.
- W2897623300 creator A5028952508 @default.
- W2897623300 creator A5029924052 @default.
- W2897623300 creator A5031076543 @default.
- W2897623300 creator A5032326710 @default.
- W2897623300 date "2018-07-01" @default.
- W2897623300 modified "2023-10-17" @default.
- W2897623300 title "Low Latency Spiking ConvNets with Restricted Output Training and False Spike Inhibition" @default.
- W2897623300 cites W101771737 @default.
- W2897623300 cites W2016708835 @default.
- W2897623300 cites W2020096355 @default.
- W2897623300 cites W2020676607 @default.
- W2897623300 cites W2076063813 @default.
- W2897623300 cites W2107433900 @default.
- W2897623300 cites W2112796928 @default.
- W2897623300 cites W2116360511 @default.
- W2897623300 cites W2159951683 @default.
- W2897623300 cites W2165639766 @default.
- W2897623300 cites W2314470091 @default.
- W2897623300 cites W2513148968 @default.
- W2897623300 cites W2513853720 @default.
- W2897623300 cites W2606722458 @default.
- W2897623300 cites W2742439472 @default.
- W2897623300 cites W2743130883 @default.
- W2897623300 cites W2775079417 @default.
- W2897623300 doi "https://doi.org/10.1109/ijcnn.2018.8489400" @default.
- W2897623300 hasPublicationYear "2018" @default.
- W2897623300 type Work @default.
- W2897623300 sameAs 2897623300 @default.
- W2897623300 citedByCount "8" @default.
- W2897623300 countsByYear W28976233002020 @default.
- W2897623300 countsByYear W28976233002021 @default.
- W2897623300 countsByYear W28976233002023 @default.
- W2897623300 crossrefType "proceedings-article" @default.
- W2897623300 hasAuthorship W2897623300A5000050947 @default.
- W2897623300 hasAuthorship W2897623300A5002976537 @default.
- W2897623300 hasAuthorship W2897623300A5028952508 @default.
- W2897623300 hasAuthorship W2897623300A5029924052 @default.
- W2897623300 hasAuthorship W2897623300A5031076543 @default.
- W2897623300 hasAuthorship W2897623300A5032326710 @default.
- W2897623300 hasConcept C11413529 @default.
- W2897623300 hasConcept C115903868 @default.
- W2897623300 hasConcept C11731999 @default.
- W2897623300 hasConcept C119857082 @default.
- W2897623300 hasConcept C153180895 @default.
- W2897623300 hasConcept C154945302 @default.
- W2897623300 hasConcept C173608175 @default.
- W2897623300 hasConcept C2781390188 @default.
- W2897623300 hasConcept C41008148 @default.
- W2897623300 hasConcept C45374587 @default.
- W2897623300 hasConcept C50644808 @default.
- W2897623300 hasConcept C68339613 @default.
- W2897623300 hasConcept C70437156 @default.
- W2897623300 hasConcept C76155785 @default.
- W2897623300 hasConcept C81363708 @default.
- W2897623300 hasConcept C82876162 @default.
- W2897623300 hasConceptScore W2897623300C11413529 @default.
- W2897623300 hasConceptScore W2897623300C115903868 @default.
- W2897623300 hasConceptScore W2897623300C11731999 @default.
- W2897623300 hasConceptScore W2897623300C119857082 @default.
- W2897623300 hasConceptScore W2897623300C153180895 @default.
- W2897623300 hasConceptScore W2897623300C154945302 @default.
- W2897623300 hasConceptScore W2897623300C173608175 @default.
- W2897623300 hasConceptScore W2897623300C2781390188 @default.
- W2897623300 hasConceptScore W2897623300C41008148 @default.
- W2897623300 hasConceptScore W2897623300C45374587 @default.
- W2897623300 hasConceptScore W2897623300C50644808 @default.
- W2897623300 hasConceptScore W2897623300C68339613 @default.
- W2897623300 hasConceptScore W2897623300C70437156 @default.
- W2897623300 hasConceptScore W2897623300C76155785 @default.
- W2897623300 hasConceptScore W2897623300C81363708 @default.
- W2897623300 hasConceptScore W2897623300C82876162 @default.
- W2897623300 hasLocation W28976233001 @default.
- W2897623300 hasOpenAccess W2897623300 @default.
- W2897623300 hasPrimaryLocation W28976233001 @default.
- W2897623300 hasRelatedWork W2291847203 @default.
- W2897623300 hasRelatedWork W2424871898 @default.
- W2897623300 hasRelatedWork W2517027266 @default.
- W2897623300 hasRelatedWork W2756241593 @default.
- W2897623300 hasRelatedWork W3004532561 @default.
- W2897623300 hasRelatedWork W3027997911 @default.
- W2897623300 hasRelatedWork W3076103167 @default.
- W2897623300 hasRelatedWork W4287776258 @default.
- W2897623300 hasRelatedWork W4295243112 @default.
- W2897623300 hasRelatedWork W4281699635 @default.
- W2897623300 isParatext "false" @default.
- W2897623300 isRetracted "false" @default.
- W2897623300 magId "2897623300" @default.
- W2897623300 workType "article" @default.