FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3120009304> ?p ?o ?g. }

• W3120009304 endingPage "461" @default.
• W3120009304 startingPage "448" @default.
• W3120009304 abstract "As one of the most promising methods in the next generation of neuromorphic systems, memristor-based spiking neural networks (SNNs) show great advantages in terms of power efficiency, integration density, and biological plausibility. However, because of the nondifferentiability of discrete spikes, it is difficult to train SNNs with gradient descent and error backpropagation online. In this article, we propose an improved training algorithm for multilayer memristive SNN (MSNN) with three methods spontaneously, supporting <italic xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>in situ</i> learning in hardware: 1) temporal order encoding is applied to generate different pulse trains in neurons; 2) a simplified homeostasis is realized by the activation state and refractory period to regulate hidden neurons spontaneously; and 3) spiking-timing-dependent plasticity (STDP) in memristive synapses is adopted to update weights <italic xmlns:mml=http://www.w3.org/1998/Math/MathML xmlns:xlink=http://www.w3.org/1999/xlink>in situ</i> . Correspondingly, we provide a circuitry example and verify it in LTSPICE. Then the MSNN is benchmarked with the MNIST data set and analyzed with visualization methods, showing better recognition accuracy (95.15%) than existing SNNs with comparable scales and bio-inspired learning rules. We also consider some nonideal effects in memristor crossbar array and peripheral circuits. Evaluation results show that the proposed MSNN is robust to finite resolution, circuit noise and writing noise; and larger network scale will help the MSNN alleviate the negative impacts of other nonideal factors, including yield and device-to-device variation. Moreover, the energy efficiency of a MSNN system is estimated to achieve 7.6TOPS/W, showing great potential in low-power applications." @default.
• W3120009304 created "2021-01-18" @default.
• W3120009304 creator A5013145898 @default.
• W3120009304 creator A5015105154 @default.
• W3120009304 creator A5023777406 @default.
• W3120009304 creator A5045152683 @default.
• W3120009304 creator A5056678009 @default.
• W3120009304 creator A5066058965 @default.
• W3120009304 creator A5077851706 @default.
• W3120009304 date "2022-06-01" @default.
• W3120009304 modified "2023-10-17" @default.
• W3120009304 title "<i>In Situ</i> Learning in Hardware Compatible Multilayer Memristive Spiking Neural Network" @default.
• W3120009304 cites W1495791788 @default.
• W3120009304 cites W1570411240 @default.
• W3120009304 cites W1645800954 @default.
• W3120009304 cites W1964288235 @default.
• W3120009304 cites W1967628726 @default.
• W3120009304 cites W1979180831 @default.
• W3120009304 cites W1995314879 @default.
• W3120009304 cites W2015861736 @default.
• W3120009304 cites W2016589492 @default.
• W3120009304 cites W2016922062 @default.
• W3120009304 cites W2026122895 @default.
• W3120009304 cites W2112181056 @default.
• W3120009304 cites W2145967274 @default.
• W3120009304 cites W2156640153 @default.
• W3120009304 cites W2162827630 @default.
• W3120009304 cites W2463288820 @default.
• W3120009304 cites W2466978667 @default.
• W3120009304 cites W2560615381 @default.
• W3120009304 cites W2610452982 @default.
• W3120009304 cites W2612375349 @default.
• W3120009304 cites W2621826044 @default.
• W3120009304 cites W2735894830 @default.
• W3120009304 cites W2765399440 @default.
• W3120009304 cites W2790636388 @default.
• W3120009304 cites W2792208628 @default.
• W3120009304 cites W2794822538 @default.
• W3120009304 cites W2804040790 @default.
• W3120009304 cites W2807750997 @default.
• W3120009304 cites W2808518417 @default.
• W3120009304 cites W2810068957 @default.
• W3120009304 cites W2884096268 @default.
• W3120009304 cites W2884136970 @default.
• W3120009304 cites W2884987480 @default.
• W3120009304 cites W2886819279 @default.
• W3120009304 cites W2896666858 @default.
• W3120009304 cites W2905533880 @default.
• W3120009304 cites W2910128286 @default.
• W3120009304 cites W2923010225 @default.
• W3120009304 cites W2963588827 @default.
• W3120009304 cites W2966081953 @default.
• W3120009304 cites W2978313259 @default.
• W3120009304 cites W2981846116 @default.
• W3120009304 cites W3000505821 @default.
• W3120009304 cites W3003217511 @default.
• W3120009304 cites W3003821665 @default.
• W3120009304 cites W3021675401 @default.
• W3120009304 doi "https://doi.org/10.1109/tcds.2021.3049487" @default.
• W3120009304 hasPublicationYear "2022" @default.
• W3120009304 type Work @default.
• W3120009304 sameAs 3120009304 @default.
• W3120009304 citedByCount "4" @default.
• W3120009304 countsByYear W31200093042021 @default.
• W3120009304 countsByYear W31200093042022 @default.
• W3120009304 countsByYear W31200093042023 @default.
• W3120009304 crossrefType "journal-article" @default.
• W3120009304 hasAuthorship W3120009304A5013145898 @default.
• W3120009304 hasAuthorship W3120009304A5015105154 @default.
• W3120009304 hasAuthorship W3120009304A5023777406 @default.
• W3120009304 hasAuthorship W3120009304A5045152683 @default.
• W3120009304 hasAuthorship W3120009304A5056678009 @default.
• W3120009304 hasAuthorship W3120009304A5066058965 @default.
• W3120009304 hasAuthorship W3120009304A5077851706 @default.
• W3120009304 hasConcept C115961682 @default.
• W3120009304 hasConcept C11731999 @default.
• W3120009304 hasConcept C127413603 @default.
• W3120009304 hasConcept C150072547 @default.
• W3120009304 hasConcept C151927369 @default.
• W3120009304 hasConcept C154945302 @default.
• W3120009304 hasConcept C190502265 @default.
• W3120009304 hasConcept C24326235 @default.
• W3120009304 hasConcept C29984679 @default.
• W3120009304 hasConcept C41008148 @default.
• W3120009304 hasConcept C50644808 @default.
• W3120009304 hasConcept C76155785 @default.
• W3120009304 hasConcept C99498987 @default.
• W3120009304 hasConceptScore W3120009304C115961682 @default.
• W3120009304 hasConceptScore W3120009304C11731999 @default.
• W3120009304 hasConceptScore W3120009304C127413603 @default.
• W3120009304 hasConceptScore W3120009304C150072547 @default.
• W3120009304 hasConceptScore W3120009304C151927369 @default.
• W3120009304 hasConceptScore W3120009304C154945302 @default.
• W3120009304 hasConceptScore W3120009304C190502265 @default.
• W3120009304 hasConceptScore W3120009304C24326235 @default.
• W3120009304 hasConceptScore W3120009304C29984679 @default.
• W3120009304 hasConceptScore W3120009304C41008148 @default.
• W3120009304 hasConceptScore W3120009304C50644808 @default.

• next