Matches in SemOpenAlex for { <https://semopenalex.org/work/W4309217996> ?p ?o ?g. }
Showing items 1 to 64 of
64
with 100 items per page.
- W4309217996 abstract "Spiking neural networks (SNNs) are well suited for resource-constrained applications as they do not need expensive multipliers. In a typical rate-encoded SNN, a series of binary spikes within a globally fixed time window is used to fire the neurons. The maximum number of spikes in this time window is also the latency of the network in performing a single inference, as well as determines the overall energy efficiency of the model. The aim of this paper is to reduce this while maintaining accuracy when converting ANNs to their equivalent SNNs. The state-of-the-art conversion schemes yield SNNs with accuracies comparable with ANNs only for large window sizes. In this paper, we start with understanding the information loss when converting from pre-existing ANN models to standard rate-encoded SNN models. From these insights, we propose a suite of novel techniques that together mitigate the information lost in the conversion, and achieve state-of-art SNN accuracies along with very low latency. Our method achieved a Top-1 SNN accuracy of 98.73% (1 time step) on the MNIST dataset, 76.38% (8 time steps) on the CIFAR-100 dataset, and 93.71% (8 time steps) on the CIFAR-10 dataset. On ImageNet, an SNN accuracy of 75.35%/79.16% was achieved with 100/200 time steps." @default.
- W4309217996 created "2022-11-24" @default.
- W4309217996 creator A5004250652 @default.
- W4309217996 creator A5023989495 @default.
- W4309217996 creator A5089103033 @default.
- W4309217996 date "2022-10-27" @default.
- W4309217996 modified "2023-09-27" @default.
- W4309217996 title "Low Latency Conversion of Artificial Neural Network Models to Rate-encoded Spiking Neural Networks" @default.
- W4309217996 doi "https://doi.org/10.48550/arxiv.2211.08410" @default.
- W4309217996 hasPublicationYear "2022" @default.
- W4309217996 type Work @default.
- W4309217996 citedByCount "0" @default.
- W4309217996 crossrefType "posted-content" @default.
- W4309217996 hasAuthorship W4309217996A5004250652 @default.
- W4309217996 hasAuthorship W4309217996A5023989495 @default.
- W4309217996 hasAuthorship W4309217996A5089103033 @default.
- W4309217996 hasBestOaLocation W43092179961 @default.
- W4309217996 hasConcept C111919701 @default.
- W4309217996 hasConcept C11731999 @default.
- W4309217996 hasConcept C153180895 @default.
- W4309217996 hasConcept C154945302 @default.
- W4309217996 hasConcept C166957645 @default.
- W4309217996 hasConcept C190502265 @default.
- W4309217996 hasConcept C2776214188 @default.
- W4309217996 hasConcept C2778751112 @default.
- W4309217996 hasConcept C2984842247 @default.
- W4309217996 hasConcept C41008148 @default.
- W4309217996 hasConcept C50644808 @default.
- W4309217996 hasConcept C76155785 @default.
- W4309217996 hasConcept C79581498 @default.
- W4309217996 hasConcept C82876162 @default.
- W4309217996 hasConcept C95457728 @default.
- W4309217996 hasConceptScore W4309217996C111919701 @default.
- W4309217996 hasConceptScore W4309217996C11731999 @default.
- W4309217996 hasConceptScore W4309217996C153180895 @default.
- W4309217996 hasConceptScore W4309217996C154945302 @default.
- W4309217996 hasConceptScore W4309217996C166957645 @default.
- W4309217996 hasConceptScore W4309217996C190502265 @default.
- W4309217996 hasConceptScore W4309217996C2776214188 @default.
- W4309217996 hasConceptScore W4309217996C2778751112 @default.
- W4309217996 hasConceptScore W4309217996C2984842247 @default.
- W4309217996 hasConceptScore W4309217996C41008148 @default.
- W4309217996 hasConceptScore W4309217996C50644808 @default.
- W4309217996 hasConceptScore W4309217996C76155785 @default.
- W4309217996 hasConceptScore W4309217996C79581498 @default.
- W4309217996 hasConceptScore W4309217996C82876162 @default.
- W4309217996 hasConceptScore W4309217996C95457728 @default.
- W4309217996 hasLocation W43092179961 @default.
- W4309217996 hasLocation W43092179962 @default.
- W4309217996 hasOpenAccess W4309217996 @default.
- W4309217996 hasPrimaryLocation W43092179961 @default.
- W4309217996 hasRelatedWork W2133796295 @default.
- W4309217996 hasRelatedWork W2742991909 @default.
- W4309217996 hasRelatedWork W2810865670 @default.
- W4309217996 hasRelatedWork W2936783136 @default.
- W4309217996 hasRelatedWork W2966649771 @default.
- W4309217996 hasRelatedWork W3042419602 @default.
- W4309217996 hasRelatedWork W4200635499 @default.
- W4309217996 hasRelatedWork W4236949603 @default.
- W4309217996 hasRelatedWork W4243928383 @default.
- W4309217996 hasRelatedWork W4281697256 @default.
- W4309217996 isParatext "false" @default.
- W4309217996 isRetracted "false" @default.
- W4309217996 workType "article" @default.