Matches in SemOpenAlex for { <https://semopenalex.org/work/W4282829327> ?p ?o ?g. }
- W4282829327 endingPage "032001" @default.
- W4282829327 startingPage "032001" @default.
- W4282829327 abstract "Abstract In the internet-of-things era, edge intelligence is critical for overcoming the communication and computing energy crisis, which is unavoidable if cloud computing is used exclusively. Memristor crossbars with in-memory computing may be suitable for realizing edge intelligence hardware. They can perform both memory and computing functions, allowing for the development of low-power computing architectures that go beyond the von Neumann computer. For implementing edge-intelligence hardware with memristor crossbars, in this paper, we review various techniques such as quantization, training, parasitic resistance correction, and low-power crossbar programming, and so on. In particular, memristor crossbars can be considered to realize quantized neural networks with binary and ternary synapses. For preventing memristor defects from degrading edge intelligence performance, chip-in-the-loop training can be useful when training memristor crossbars. Another undesirable effect in memristor crossbars is parasitic resistances such as source, line, and neuron resistance, which worsens as crossbar size increases. Various circuit and software techniques can compensate for parasitic resistances like source, line, and neuron resistance. Finally, we discuss an energy-efficient programming method for updating synaptic weights in memristor crossbars, which is needed for learning the edge devices." @default.
- W4282829327 created "2022-06-15" @default.
- W4282829327 creator A5004841618 @default.
- W4282829327 creator A5063226833 @default.
- W4282829327 creator A5065993014 @default.
- W4282829327 creator A5076797879 @default.
- W4282829327 creator A5089653706 @default.
- W4282829327 date "2022-07-07" @default.
- W4282829327 modified "2023-10-14" @default.
- W4282829327 title "Quantization, training, parasitic resistance correction, and programming techniques of memristor-crossbar neural networks for edge intelligence" @default.
- W4282829327 cites W1995314879 @default.
- W4282829327 cites W2000747118 @default.
- W4282829327 cites W2016922062 @default.
- W4282829327 cites W2021383442 @default.
- W4282829327 cites W2033592371 @default.
- W4282829327 cites W2036687738 @default.
- W4282829327 cites W2060897054 @default.
- W4282829327 cites W2085862958 @default.
- W4282829327 cites W2112181056 @default.
- W4282829327 cites W2135210684 @default.
- W4282829327 cites W2144085790 @default.
- W4282829327 cites W2284773887 @default.
- W4282829327 cites W2342792048 @default.
- W4282829327 cites W2485720998 @default.
- W4282829327 cites W2559718113 @default.
- W4282829327 cites W2564810971 @default.
- W4282829327 cites W2587921983 @default.
- W4282829327 cites W2756816131 @default.
- W4282829327 cites W2769561581 @default.
- W4282829327 cites W2769815858 @default.
- W4282829327 cites W2775771159 @default.
- W4282829327 cites W2782791387 @default.
- W4282829327 cites W2786027963 @default.
- W4282829327 cites W2798539958 @default.
- W4282829327 cites W2807750997 @default.
- W4282829327 cites W2810068957 @default.
- W4282829327 cites W2888961626 @default.
- W4282829327 cites W2890653437 @default.
- W4282829327 cites W2894293853 @default.
- W4282829327 cites W2900162849 @default.
- W4282829327 cites W2914770420 @default.
- W4282829327 cites W2916560903 @default.
- W4282829327 cites W2936325645 @default.
- W4282829327 cites W2942624945 @default.
- W4282829327 cites W2943313495 @default.
- W4282829327 cites W2946641467 @default.
- W4282829327 cites W2950865323 @default.
- W4282829327 cites W2955103925 @default.
- W4282829327 cites W2957701640 @default.
- W4282829327 cites W2966740705 @default.
- W4282829327 cites W2971544482 @default.
- W4282829327 cites W2993457717 @default.
- W4282829327 cites W3003821665 @default.
- W4282829327 cites W3008515144 @default.
- W4282829327 cites W3012190633 @default.
- W4282829327 cites W3015724253 @default.
- W4282829327 cites W3042879788 @default.
- W4282829327 cites W3081302630 @default.
- W4282829327 cites W3083618915 @default.
- W4282829327 cites W3090974145 @default.
- W4282829327 cites W3106392217 @default.
- W4282829327 cites W3113256013 @default.
- W4282829327 cites W3113288818 @default.
- W4282829327 cites W3174261480 @default.
- W4282829327 cites W3186511633 @default.
- W4282829327 cites W3203249428 @default.
- W4282829327 cites W3206989388 @default.
- W4282829327 doi "https://doi.org/10.1088/2634-4386/ac781a" @default.
- W4282829327 hasPublicationYear "2022" @default.
- W4282829327 type Work @default.
- W4282829327 citedByCount "3" @default.
- W4282829327 countsByYear W42828293272022 @default.
- W4282829327 countsByYear W42828293272023 @default.
- W4282829327 crossrefType "journal-article" @default.
- W4282829327 hasAuthorship W4282829327A5004841618 @default.
- W4282829327 hasAuthorship W4282829327A5063226833 @default.
- W4282829327 hasAuthorship W4282829327A5065993014 @default.
- W4282829327 hasAuthorship W4282829327A5076797879 @default.
- W4282829327 hasAuthorship W4282829327A5089653706 @default.
- W4282829327 hasBestOaLocation W42828293271 @default.
- W4282829327 hasConcept C111919701 @default.
- W4282829327 hasConcept C11413529 @default.
- W4282829327 hasConcept C118524514 @default.
- W4282829327 hasConcept C119599485 @default.
- W4282829327 hasConcept C123593499 @default.
- W4282829327 hasConcept C127413603 @default.
- W4282829327 hasConcept C150072547 @default.
- W4282829327 hasConcept C151927369 @default.
- W4282829327 hasConcept C154945302 @default.
- W4282829327 hasConcept C162307627 @default.
- W4282829327 hasConcept C164120249 @default.
- W4282829327 hasConcept C165801399 @default.
- W4282829327 hasConcept C173608175 @default.
- W4282829327 hasConcept C182019814 @default.
- W4282829327 hasConcept C1895703 @default.
- W4282829327 hasConcept C194222762 @default.
- W4282829327 hasConcept C23123220 @default.
- W4282829327 hasConcept C24326235 @default.