• DocumentCode
    3547331
  • Title

    An adaptive analog synapse circuit that implements the least-mean-square learning rule

  • Author

    Srinivasan, Venkatesh ; Dugger, Jeff ; Hasler, Paul

  • Author_Institution
    Sch. of Electr. & Comput. Eng., Georgia Inst. of Technol., Atlanta, GA, USA
  • fYear
    2005
  • fDate
    23-26 May 2005
  • Firstpage
    4441
  • Abstract
    In this paper a compact adaptive analog synapse circuit that implements the least-mean-square (LMS) learning rule is described. Basic simulation results demonstrate the LMS learning rule in the proposed circuit. An adaptive linear combiner that uses the proposed synapse is shown to learn a square wave that matches closely with the desired target. Issues of weight decay and its implications to the design of the synapse circuit are presented as well. The synapse is designed in a 0.5 μm CMOS technology.
  • Keywords
    CMOS integrated circuits; adaptive filters; learning (artificial intelligence); least mean squares methods; neural nets; 0.5 micron; CMOS technology; LMS learning rule; adaptive analog synapse circuit; adaptive linear combiner; least-mean-square learning rule; weight decay; Adaptation model; Adaptive filters; Adaptive systems; CMOS technology; Circuit simulation; Computational modeling; Least squares approximation; Nonvolatile memory; Transversal filters; Tunneling;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Circuits and Systems, 2005. ISCAS 2005. IEEE International Symposium on
  • Print_ISBN
    0-7803-8834-8
  • Type

    conf

  • DOI
    10.1109/ISCAS.2005.1465617
  • Filename
    1465617