FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3100731286> ?p ?o ?g. }

• W3100731286 abstract "To tackle real-world challenges, deep and complex neural networks are generally used with a massive number of parameters, which require large memory size, extensive computational operations, and high energy consumption in neuromorphic hardware systems. In this work, we propose an unsupervised online adaptive weight pruning method that dynamically removes noncritical weights from a spiking neural network (SNN) to reduce network complexity and improve energy efficiency. The adaptive pruning method explores neural dynamics and firing activity of SNNs and adapts the pruning threshold over time and neurons during training. The proposed adaption scheme allows the network to effectively identify critical weights associated with each neuron by changing the pruning threshold dynamically over time and neurons. It balances the connection strength of neurons with the previous layer with adaptive thresholds and prevents weak neurons from failure after pruning. We also evaluated improvement in the energy efficiency of SNNs with our method by computing synaptic operations (SOPs). Simulation results and detailed analyses have revealed that applying adaption in the pruning threshold can significantly improve network performance and reduce the number of SOPs. The pruned SNN with 800 excitatory neurons can achieve a 30% reduction in SOPs during training and a 55% reduction during inference, with only 0.44% accuracy loss on MNIST dataset. Compared with a previously reported online soft pruning method, the proposed adaptive pruning method shows 3.33% higher classification accuracy and 67% more reduction in synaptic operations. The effectiveness of our method was confirmed on different datasets and for different network sizes. Our evaluation showed that the implementation overhead of the adaptive method regarding speed, area and energy is negligible in the network. Therefore, this work offers a promising solution for effective network compression and building highly energy-efficient neuromorphic systems in real-time applications." @default.
• W3100731286 created "2020-11-23" @default.
• W3100731286 creator A5003795739 @default.
• W3100731286 creator A5012920132 @default.
• W3100731286 creator A5013660772 @default.
• W3100731286 creator A5068089858 @default.
• W3100731286 creator A5084829415 @default.
• W3100731286 date "2020-11-12" @default.
• W3100731286 modified "2023-09-27" @default.
• W3100731286 title "Unsupervised Adaptive Weight Pruning for Energy-Efficient Neuromorphic Systems" @default.
• W3100731286 cites W1570411240 @default.
• W3100731286 cites W1975412204 @default.
• W3100731286 cites W1981445810 @default.
• W3100731286 cites W2008008156 @default.
• W3100731286 cites W2048812218 @default.
• W3100731286 cites W2109596721 @default.
• W3100731286 cites W2112796928 @default.
• W3100731286 cites W2138913040 @default.
• W3100731286 cites W2159951683 @default.
• W3100731286 cites W2163630896 @default.
• W3100731286 cites W2276892413 @default.
• W3100731286 cites W2473481124 @default.
• W3100731286 cites W2771111398 @default.
• W3100731286 cites W2783525259 @default.
• W3100731286 cites W2903711666 @default.
• W3100731286 cites W2940297489 @default.
• W3100731286 cites W2942231644 @default.
• W3100731286 cites W2945098486 @default.
• W3100731286 cites W2963966976 @default.
• W3100731286 cites W2964010909 @default.
• W3100731286 cites W3012223454 @default.
• W3100731286 cites W3037245120 @default.
• W3100731286 doi "https://doi.org/10.3389/fnins.2020.598876" @default.
• W3100731286 hasPubMedCentralId "https://www.ncbi.nlm.nih.gov/pmc/articles/7689062" @default.
• W3100731286 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/33281549" @default.
• W3100731286 hasPublicationYear "2020" @default.
• W3100731286 type Work @default.
• W3100731286 sameAs 3100731286 @default.
• W3100731286 citedByCount "10" @default.
• W3100731286 countsByYear W31007312862021 @default.
• W3100731286 countsByYear W31007312862022 @default.
• W3100731286 countsByYear W31007312862023 @default.
• W3100731286 crossrefType "journal-article" @default.
• W3100731286 hasAuthorship W3100731286A5003795739 @default.
• W3100731286 hasAuthorship W3100731286A5012920132 @default.
• W3100731286 hasAuthorship W3100731286A5013660772 @default.
• W3100731286 hasAuthorship W3100731286A5068089858 @default.
• W3100731286 hasAuthorship W3100731286A5084829415 @default.
• W3100731286 hasBestOaLocation W31007312861 @default.
• W3100731286 hasConcept C105795698 @default.
• W3100731286 hasConcept C108010975 @default.
• W3100731286 hasConcept C111335779 @default.
• W3100731286 hasConcept C11731999 @default.
• W3100731286 hasConcept C119857082 @default.
• W3100731286 hasConcept C151927369 @default.
• W3100731286 hasConcept C154945302 @default.
• W3100731286 hasConcept C186370098 @default.
• W3100731286 hasConcept C190502265 @default.
• W3100731286 hasConcept C2524010 @default.
• W3100731286 hasConcept C2776214188 @default.
• W3100731286 hasConcept C33923547 @default.
• W3100731286 hasConcept C41008148 @default.
• W3100731286 hasConcept C50644808 @default.
• W3100731286 hasConcept C6557445 @default.
• W3100731286 hasConcept C86803240 @default.
• W3100731286 hasConceptScore W3100731286C105795698 @default.
• W3100731286 hasConceptScore W3100731286C108010975 @default.
• W3100731286 hasConceptScore W3100731286C111335779 @default.
• W3100731286 hasConceptScore W3100731286C11731999 @default.
• W3100731286 hasConceptScore W3100731286C119857082 @default.
• W3100731286 hasConceptScore W3100731286C151927369 @default.
• W3100731286 hasConceptScore W3100731286C154945302 @default.
• W3100731286 hasConceptScore W3100731286C186370098 @default.
• W3100731286 hasConceptScore W3100731286C190502265 @default.
• W3100731286 hasConceptScore W3100731286C2524010 @default.
• W3100731286 hasConceptScore W3100731286C2776214188 @default.
• W3100731286 hasConceptScore W3100731286C33923547 @default.
• W3100731286 hasConceptScore W3100731286C41008148 @default.
• W3100731286 hasConceptScore W3100731286C50644808 @default.
• W3100731286 hasConceptScore W3100731286C6557445 @default.
• W3100731286 hasConceptScore W3100731286C86803240 @default.
• W3100731286 hasFunder F4320322320 @default.
• W3100731286 hasLocation W31007312861 @default.
• W3100731286 hasLocation W31007312862 @default.
• W3100731286 hasLocation W31007312863 @default.
• W3100731286 hasLocation W31007312864 @default.
• W3100731286 hasOpenAccess W3100731286 @default.
• W3100731286 hasPrimaryLocation W31007312861 @default.
• W3100731286 hasRelatedWork W2809732489 @default.
• W3100731286 hasRelatedWork W3023361272 @default.
• W3100731286 hasRelatedWork W3035640865 @default.
• W3100731286 hasRelatedWork W3100731286 @default.
• W3100731286 hasRelatedWork W3117246098 @default.
• W3100731286 hasRelatedWork W3161396968 @default.
• W3100731286 hasRelatedWork W3214713078 @default.
• W3100731286 hasRelatedWork W4287780255 @default.
• W3100731286 hasRelatedWork W4385334613 @default.
• W3100731286 hasRelatedWork W4281699635 @default.
• W3100731286 hasVolume "14" @default.
• W3100731286 isParatext "false" @default.

• next