Matches in SemOpenAlex for { <https://semopenalex.org/work/W4367000460> ?p ?o ?g. }
Showing items 1 to 71 of
71
with 100 items per page.
- W4367000460 abstract "Biologically inspired Spiking Neural Networks (SNNs) have attracted significant attention for their ability to provide extremely energy-efficient machine intelligence through event-driven operation and sparse activities. As artificial intelligence (AI) becomes ever more democratized, there is an increasing need to execute SNN models on edge devices. Existing works adopt weight pruning to reduce SNN model size and accelerate inference. However, these methods mainly focus on how to obtain a sparse model for efficient inference, rather than training efficiency. To overcome these drawbacks, in this paper, we propose a Neurogenesis Dynamics-inspired Spiking Neural Network training acceleration framework, NDSNN. Our framework is computational efficient and trains a model from scratch with dynamic sparsity without sacrificing model fidelity. Specifically, we design a new drop-and-grow strategy with decreasing number of non-zero weights, to maintain extreme high sparsity and high accuracy. We evaluate NDSNN using VGG-16 and ResNet-19 on CIFAR-10, CIFAR-100 and TinyImageNet. Experimental results show that NDSNN achieves up to 20.52% improvement in accuracy on Tiny-ImageNet using ResNet-19 (with a sparsity of 99%) as compared to other SOTA methods (e.g., Lottery Ticket Hypothesis (LTH), SET-SNN, RigL-SNN). In addition, the training cost of NDSNN is only 40.89% of the LTH training cost on ResNet-19 and 31.35% of the LTH training cost on VGG-16 on CIFAR-10." @default.
- W4367000460 created "2023-04-27" @default.
- W4367000460 creator A5008547311 @default.
- W4367000460 creator A5030060072 @default.
- W4367000460 creator A5042449700 @default.
- W4367000460 creator A5051105499 @default.
- W4367000460 creator A5051195699 @default.
- W4367000460 creator A5053352602 @default.
- W4367000460 creator A5060399455 @default.
- W4367000460 creator A5067226050 @default.
- W4367000460 creator A5068433690 @default.
- W4367000460 creator A5087003069 @default.
- W4367000460 date "2023-04-24" @default.
- W4367000460 modified "2023-10-16" @default.
- W4367000460 title "Neurogenesis Dynamics-inspired Spiking Neural Network Training Acceleration" @default.
- W4367000460 doi "https://doi.org/10.48550/arxiv.2304.12214" @default.
- W4367000460 hasPublicationYear "2023" @default.
- W4367000460 type Work @default.
- W4367000460 citedByCount "0" @default.
- W4367000460 crossrefType "posted-content" @default.
- W4367000460 hasAuthorship W4367000460A5008547311 @default.
- W4367000460 hasAuthorship W4367000460A5030060072 @default.
- W4367000460 hasAuthorship W4367000460A5042449700 @default.
- W4367000460 hasAuthorship W4367000460A5051105499 @default.
- W4367000460 hasAuthorship W4367000460A5051195699 @default.
- W4367000460 hasAuthorship W4367000460A5053352602 @default.
- W4367000460 hasAuthorship W4367000460A5060399455 @default.
- W4367000460 hasAuthorship W4367000460A5067226050 @default.
- W4367000460 hasAuthorship W4367000460A5068433690 @default.
- W4367000460 hasAuthorship W4367000460A5087003069 @default.
- W4367000460 hasBestOaLocation W43670004601 @default.
- W4367000460 hasConcept C108010975 @default.
- W4367000460 hasConcept C115537543 @default.
- W4367000460 hasConcept C11731999 @default.
- W4367000460 hasConcept C119857082 @default.
- W4367000460 hasConcept C133588205 @default.
- W4367000460 hasConcept C154945302 @default.
- W4367000460 hasConcept C173608175 @default.
- W4367000460 hasConcept C2776214188 @default.
- W4367000460 hasConcept C41008148 @default.
- W4367000460 hasConcept C50644808 @default.
- W4367000460 hasConcept C6557445 @default.
- W4367000460 hasConcept C86803240 @default.
- W4367000460 hasConceptScore W4367000460C108010975 @default.
- W4367000460 hasConceptScore W4367000460C115537543 @default.
- W4367000460 hasConceptScore W4367000460C11731999 @default.
- W4367000460 hasConceptScore W4367000460C119857082 @default.
- W4367000460 hasConceptScore W4367000460C133588205 @default.
- W4367000460 hasConceptScore W4367000460C154945302 @default.
- W4367000460 hasConceptScore W4367000460C173608175 @default.
- W4367000460 hasConceptScore W4367000460C2776214188 @default.
- W4367000460 hasConceptScore W4367000460C41008148 @default.
- W4367000460 hasConceptScore W4367000460C50644808 @default.
- W4367000460 hasConceptScore W4367000460C6557445 @default.
- W4367000460 hasConceptScore W4367000460C86803240 @default.
- W4367000460 hasLocation W43670004601 @default.
- W4367000460 hasOpenAccess W4367000460 @default.
- W4367000460 hasPrimaryLocation W43670004601 @default.
- W4367000460 hasRelatedWork W1554521794 @default.
- W4367000460 hasRelatedWork W1575002289 @default.
- W4367000460 hasRelatedWork W2347422207 @default.
- W4367000460 hasRelatedWork W2353430311 @default.
- W4367000460 hasRelatedWork W2382239615 @default.
- W4367000460 hasRelatedWork W2511279186 @default.
- W4367000460 hasRelatedWork W2963058055 @default.
- W4367000460 hasRelatedWork W3199608561 @default.
- W4367000460 hasRelatedWork W1629725936 @default.
- W4367000460 hasRelatedWork W2528885285 @default.
- W4367000460 isParatext "false" @default.
- W4367000460 isRetracted "false" @default.
- W4367000460 workType "article" @default.