• DocumentCode
    3629573
  • Title

    Experimental results of CMOS-implemented conscience mechanism applied for WTA networks

  • Author

    Tomasz Talaska;Rafal Dlugosz;Jakub Dalecki;Witold Pedrycz;Ryszard Wojtyna

  • Author_Institution
    University of Technology and Life Sciences, Faculty of Telecommunication and Electrical Engineering, ul. Kaliskiego 7, 85-791 Bydgoszcz, Poland
  • fYear
    2008
  • Firstpage
    101
  • Lastpage
    104
  • Abstract
    This paper presents experimental results demonstrating a positive effect of using conscience mechanism in on chip learning Kohonen networks. Application of this mechanism allows to eliminate so called ldquodead neuronsrdquo, which do not take part in the competition during the learning phase, thus increasing the quantization error of the network. Both Matlab simulations and measurement results are shown.
  • Keywords
    "Neurons","Neural networks","CMOS technology","Neural network hardware","Network-on-a-chip","Silicon","Electronic mail","Quantization","Computational modeling","Semiconductor device measurement"
  • Publisher
    ieee
  • Conference_Titel
    Signals and Electronic Systems, 2008. ICSES ´08. International Conference on
  • Print_ISBN
    978-83-88309-47-2
  • Type

    conf

  • DOI
    10.1109/ICSES.2008.4673368
  • Filename
    4673368
    • Link To Document
    https://search.isc.ac/dl/search/defaultta.aspx?DTC=49&DC=3629573