• DocumentCode
    808087
  • Title

    A refreshable analog VLSI neural network chip with 400 neurons and 40 K synapses

  • Author

    Arima, Yutaka ; Murasaki, Mitsuhiro ; Yamada, Tsuyoshi ; Maeda, Atsushi ; Shinohara, Hirofumi

  • Author_Institution
    Mitsubishi Electr. Corp., Hyogo, Japan
  • Volume
    27
  • Issue
    12
  • fYear
    1992
  • fDate
    12/1/1992 12:00:00 AM
  • Firstpage
    1854
  • Lastpage
    1861
  • Abstract
    An on-chip learning neural network LSI circuit that can refresh the analog storage synaptic weights located on a chip is described. The chip integrates 400 neurons and 40000 synapses with a 0.8-μm double-poly double-metal CMOS technology. This device stores learned information by repeating the refresh process at 200-ms intervals
  • Keywords
    CMOS integrated circuits; VLSI; analogue processing circuits; learning (artificial intelligence); neural chips; 0.8 micron; analog storage synaptic weights; double-poly double-metal CMOS technology; neurons; on-chip learning neural network LSI circuit; refreshable analog VLSI neural network chip; synapses; CMOS technology; Capacitors; Circuits; Computer networks; Large scale integration; Large-scale systems; Network-on-a-chip; Neural networks; Neurons; Very large scale integration;
  • fLanguage
    English
  • Journal_Title
    Solid-State Circuits, IEEE Journal of
  • Publisher
    ieee
  • ISSN
    0018-9200
  • Type

    jour

  • DOI
    10.1109/4.173115
  • Filename
    173115