• DocumentCode
    908095
  • Title

    `Winner-take-all´ circuit for neurocomputing applications

  • Author

    Perfetti, R.

  • Author_Institution
    INFO-COM Dept., Roma, Italy
  • Volume
    137
  • Issue
    5
  • fYear
    1990
  • fDate
    10/1/1990 12:00:00 AM
  • Firstpage
    353
  • Lastpage
    359
  • Abstract
    The author proposes a parallel analogue implementation of the `winner-take-all´ function. It can be utilised in the hardware realisation of neurocomputer systems based on competitive learning, which find application in areas such as pattern classification, vector quantisation and image recognition. Analytic results are derived by means of the piecewise-linear method in the case of two variables and SPICE simulation results are presented in the general case. It is shown that the circuit exhibits some ad hoc properties that make it well suited for this kind of application
  • Keywords
    analogue computer circuits; brain models; circuit analysis computing; learning systems; neural nets; nonlinear network synthesis; piecewise-linear techniques; SPICE simulation; ad hoc properties; image recognition; neurocomputer systems; parallel analogue implementation; pattern classification; piecewise-linear method; vector quantisation; winner-take-all circuits;
  • fLanguage
    English
  • Journal_Title
    Circuits, Devices and Systems, IEE Proceedings G
  • Publisher
    iet
  • ISSN
    0956-3768
  • Type

    jour

  • Filename
    217103