Matches in SemOpenAlex for { <https://semopenalex.org/work/W4297889504> ?p ?o ?g. }
Showing items 1 to 76 of
76
with 100 items per page.
- W4297889504 abstract "Spiking Neural Networks (SNNs) use spatio-temporal spike patterns to represent and transmit information, which is not only biologically realistic but also suitable for ultra-low-power event-driven neuromorphic implementation. Motivated by the success of deep learning, the study of Deep Spiking Neural Networks (DeepSNNs) provides promising directions for artificial intelligence applications. However, training of DeepSNNs is not straightforward because the well-studied error back-propagation (BP) algorithm is not directly applicable. In this paper, we first establish an understanding as to why error back-propagation does not work well in DeepSNNs. To address this problem, we propose a simple yet efficient Rectified Linear Postsynaptic Potential function (ReL-PSP) for spiking neurons and propose a Spike-Timing-Dependent Back-Propagation (STDBP) learning algorithm for DeepSNNs. In STDBP algorithm, the timing of individual spikes is used to convey information (temporal coding), and learning (back-propagation) is performed based on spike timing in an event-driven manner. Our experimental results show that the proposed learning algorithm achieves state-of-the-art classification accuracy in single spike time based learning algorithms of DeepSNNs. Furthermore, by utilizing the trained model parameters obtained from the proposed STDBP learning algorithm, we demonstrate the ultra-low-power inference operations on a recently proposed neuromorphic inference accelerator. Experimental results show that the neuromorphic hardware consumes 0.751~mW of the total power consumption and achieves a low latency of 47.71~ms to classify an image from the MNIST dataset. Overall, this work investigates the contribution of spike timing dynamics to information encoding, synaptic plasticity and decision making, providing a new perspective to design of future DeepSNNs and neuromorphic hardware systems." @default.
- W4297889504 created "2022-10-01" @default.
- W4297889504 creator A5004148006 @default.
- W4297889504 creator A5017334889 @default.
- W4297889504 creator A5025628748 @default.
- W4297889504 creator A5031523730 @default.
- W4297889504 creator A5031886937 @default.
- W4297889504 creator A5032315743 @default.
- W4297889504 creator A5032690182 @default.
- W4297889504 creator A5042364148 @default.
- W4297889504 creator A5066851412 @default.
- W4297889504 creator A5079469365 @default.
- W4297889504 creator A5069683581 @default.
- W4297889504 date "2020-03-26" @default.
- W4297889504 modified "2023-09-26" @default.
- W4297889504 title "Rectified Linear Postsynaptic Potential Function for Backpropagation in Deep Spiking Neural Networks" @default.
- W4297889504 doi "https://doi.org/10.48550/arxiv.2003.11837" @default.
- W4297889504 hasPublicationYear "2020" @default.
- W4297889504 type Work @default.
- W4297889504 citedByCount "0" @default.
- W4297889504 crossrefType "posted-content" @default.
- W4297889504 hasAuthorship W4297889504A5004148006 @default.
- W4297889504 hasAuthorship W4297889504A5017334889 @default.
- W4297889504 hasAuthorship W4297889504A5025628748 @default.
- W4297889504 hasAuthorship W4297889504A5031523730 @default.
- W4297889504 hasAuthorship W4297889504A5031886937 @default.
- W4297889504 hasAuthorship W4297889504A5032315743 @default.
- W4297889504 hasAuthorship W4297889504A5032690182 @default.
- W4297889504 hasAuthorship W4297889504A5042364148 @default.
- W4297889504 hasAuthorship W4297889504A5066851412 @default.
- W4297889504 hasAuthorship W4297889504A5069683581 @default.
- W4297889504 hasAuthorship W4297889504A5079469365 @default.
- W4297889504 hasBestOaLocation W42978895041 @default.
- W4297889504 hasConcept C108583219 @default.
- W4297889504 hasConcept C115903868 @default.
- W4297889504 hasConcept C11731999 @default.
- W4297889504 hasConcept C119857082 @default.
- W4297889504 hasConcept C151927369 @default.
- W4297889504 hasConcept C153180895 @default.
- W4297889504 hasConcept C154945302 @default.
- W4297889504 hasConcept C155032097 @default.
- W4297889504 hasConcept C190502265 @default.
- W4297889504 hasConcept C2776214188 @default.
- W4297889504 hasConcept C2781390188 @default.
- W4297889504 hasConcept C41008148 @default.
- W4297889504 hasConcept C50644808 @default.
- W4297889504 hasConceptScore W4297889504C108583219 @default.
- W4297889504 hasConceptScore W4297889504C115903868 @default.
- W4297889504 hasConceptScore W4297889504C11731999 @default.
- W4297889504 hasConceptScore W4297889504C119857082 @default.
- W4297889504 hasConceptScore W4297889504C151927369 @default.
- W4297889504 hasConceptScore W4297889504C153180895 @default.
- W4297889504 hasConceptScore W4297889504C154945302 @default.
- W4297889504 hasConceptScore W4297889504C155032097 @default.
- W4297889504 hasConceptScore W4297889504C190502265 @default.
- W4297889504 hasConceptScore W4297889504C2776214188 @default.
- W4297889504 hasConceptScore W4297889504C2781390188 @default.
- W4297889504 hasConceptScore W4297889504C41008148 @default.
- W4297889504 hasConceptScore W4297889504C50644808 @default.
- W4297889504 hasLocation W42978895041 @default.
- W4297889504 hasLocation W42978895042 @default.
- W4297889504 hasOpenAccess W4297889504 @default.
- W4297889504 hasPrimaryLocation W42978895041 @default.
- W4297889504 hasRelatedWork W2921035465 @default.
- W4297889504 hasRelatedWork W2947957795 @default.
- W4297889504 hasRelatedWork W2994834230 @default.
- W4297889504 hasRelatedWork W3156786002 @default.
- W4297889504 hasRelatedWork W3161396968 @default.
- W4297889504 hasRelatedWork W3202412105 @default.
- W4297889504 hasRelatedWork W3214713078 @default.
- W4297889504 hasRelatedWork W4287639722 @default.
- W4297889504 hasRelatedWork W4288337556 @default.
- W4297889504 hasRelatedWork W4288417961 @default.
- W4297889504 isParatext "false" @default.
- W4297889504 isRetracted "false" @default.
- W4297889504 workType "article" @default.