FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4366279870> ?p ?o ?g. }

• W4366279870 abstract "Abstract Spiking neural networks (SNNs) are more energy- and resource-efficient than artificial neural networks (ANNs). However, supervised SNN learning is a challenging task due to non-differentiability of spikes and computation of complex terms. Moreover, the design of SNN learning engines is not an easy task due to limited hardware resources and tight energy constraints. In this article, a novel hardware-efficient SNN back-propagation scheme that offers fast convergence is proposed. The learning scheme does not require any complex operation such as error normalization and weight-threshold balancing, and can achieve an accuracy of around 97.5% on MNIST dataset using only 158,800 synapses. The multiplier-less inference engine trained using the proposed hard sigmoid SNN training (HaSiST) scheme can operate at a frequency of 135 MHz and consumes only 1.03 slice registers per synapse, 2.8 slice look-up tables, and can infer about 0.03 $$times {varvec{10}}^{varvec{9}}$$ <mml:math xmlns:mml=http://www.w3.org/1998/Math/MathML> <mml:mrow> <mml:mo>×</mml:mo> <mml:msup> <mml:mrow> <mml:mrow> <mml:mn>10</mml:mn> </mml:mrow> </mml:mrow> <mml:mrow> <mml:mn>9</mml:mn> </mml:mrow> </mml:msup> </mml:mrow> </mml:math> features in a second, equivalent to 9.44 giga synaptic operations per second (GSOPS). The article also presents a high-speed, cost-efficient SNN training engine that consumes only 2.63 slice registers per synapse, 37.84 slice look-up tables per synapse, and can operate at a maximum computational frequency of around 50 MHz on a Virtex 6 FPGA." @default.
• W4366279870 created "2023-04-20" @default.
• W4366279870 creator A5032694978 @default.
• W4366279870 creator A5037922594 @default.
• W4366279870 creator A5048442222 @default.
• W4366279870 date "2023-04-18" @default.
• W4366279870 modified "2023-09-30" @default.
• W4366279870 title "A low cost neuromorphic learning engine based on a high performance supervised SNN learning algorithm" @default.
• W4366279870 cites W1474440726 @default.
• W4366279870 cites W1570411240 @default.
• W4366279870 cites W1645800954 @default.
• W4366279870 cites W2007339694 @default.
• W4366279870 cites W2029056405 @default.
• W4366279870 cites W2080333131 @default.
• W4366279870 cites W2088192327 @default.
• W4366279870 cites W2102397476 @default.
• W4366279870 cites W2115831804 @default.
• W4366279870 cites W2138913040 @default.
• W4366279870 cites W2207307457 @default.
• W4366279870 cites W2528596247 @default.
• W4366279870 cites W2569813014 @default.
• W4366279870 cites W2571808120 @default.
• W4366279870 cites W2621826044 @default.
• W4366279870 cites W2735633774 @default.
• W4366279870 cites W2775079417 @default.
• W4366279870 cites W2792794651 @default.
• W4366279870 cites W2799448654 @default.
• W4366279870 cites W2800329759 @default.
• W4366279870 cites W2802512292 @default.
• W4366279870 cites W2896097393 @default.
• W4366279870 cites W2898323475 @default.
• W4366279870 cites W2919634026 @default.
• W4366279870 cites W2959065917 @default.
• W4366279870 cites W2963089565 @default.
• W4366279870 cites W2963206832 @default.
• W4366279870 cites W2963817554 @default.
• W4366279870 cites W2966081953 @default.
• W4366279870 cites W2978859535 @default.
• W4366279870 cites W2988481835 @default.
• W4366279870 cites W2992437543 @default.
• W4366279870 cites W2998119008 @default.
• W4366279870 cites W3015205410 @default.
• W4366279870 cites W3023719681 @default.
• W4366279870 cites W3033007178 @default.
• W4366279870 cites W3037639070 @default.
• W4366279870 cites W3046342815 @default.
• W4366279870 cites W3047010826 @default.
• W4366279870 cites W3096775547 @default.
• W4366279870 cites W3118577024 @default.
• W4366279870 cites W3124716720 @default.
• W4366279870 cites W3134747453 @default.
• W4366279870 cites W3136383255 @default.
• W4366279870 cites W3158377181 @default.
• W4366279870 cites W3190686567 @default.
• W4366279870 cites W3196680848 @default.
• W4366279870 cites W3201424517 @default.
• W4366279870 cites W3212354845 @default.
• W4366279870 cites W4229069445 @default.
• W4366279870 cites W4295832479 @default.
• W4366279870 doi "https://doi.org/10.1038/s41598-023-32120-7" @default.
• W4366279870 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37072443" @default.
• W4366279870 hasPublicationYear "2023" @default.
• W4366279870 type Work @default.
• W4366279870 citedByCount "1" @default.
• W4366279870 countsByYear W43662798702023 @default.
• W4366279870 crossrefType "journal-article" @default.
• W4366279870 hasAuthorship W4366279870A5032694978 @default.
• W4366279870 hasAuthorship W4366279870A5037922594 @default.
• W4366279870 hasAuthorship W4366279870A5048442222 @default.
• W4366279870 hasBestOaLocation W43662798701 @default.
• W4366279870 hasConcept C11413529 @default.
• W4366279870 hasConcept C119857082 @default.
• W4366279870 hasConcept C154945302 @default.
• W4366279870 hasConcept C190502265 @default.
• W4366279870 hasConcept C41008148 @default.
• W4366279870 hasConcept C50644808 @default.
• W4366279870 hasConceptScore W4366279870C11413529 @default.
• W4366279870 hasConceptScore W4366279870C119857082 @default.
• W4366279870 hasConceptScore W4366279870C154945302 @default.
• W4366279870 hasConceptScore W4366279870C190502265 @default.
• W4366279870 hasConceptScore W4366279870C41008148 @default.
• W4366279870 hasConceptScore W4366279870C50644808 @default.
• W4366279870 hasIssue "1" @default.
• W4366279870 hasLocation W43662798701 @default.
• W4366279870 hasLocation W43662798702 @default.
• W4366279870 hasLocation W43662798703 @default.
• W4366279870 hasOpenAccess W4366279870 @default.
• W4366279870 hasPrimaryLocation W43662798701 @default.
• W4366279870 hasRelatedWork W2276478028 @default.
• W4366279870 hasRelatedWork W2597787948 @default.
• W4366279870 hasRelatedWork W2908596665 @default.
• W4366279870 hasRelatedWork W2936783136 @default.
• W4366279870 hasRelatedWork W2947175736 @default.
• W4366279870 hasRelatedWork W2961085424 @default.
• W4366279870 hasRelatedWork W3011385017 @default.
• W4366279870 hasRelatedWork W4286629047 @default.
• W4366279870 hasRelatedWork W4306674287 @default.
• W4366279870 hasRelatedWork W4224009465 @default.
• W4366279870 hasVolume "13" @default.
• W4366279870 isParatext "false" @default.

• next