• DocumentCode
    1545073
  • Title

    A dynamic K-winners-take-all neural network

  • Author

    Yang, Jar-Ferr ; Chen, Chi-Ming

  • Author_Institution
    Dept. of Electr. Eng., Cheng Kung Univ., Tainan, Taiwan
  • Volume
    27
  • Issue
    3
  • fYear
    1997
  • fDate
    6/1/1997 12:00:00 AM
  • Firstpage
    523
  • Lastpage
    526
  • Abstract
    In this paper, a dynamic K-winners-take-all (KWTA) neural network, which can quickly identify the K-winning neurons whose activations are larger than the remaining ones, is proposed and analyzed. For N competitors, the proposed KWTA network is composed of N feedforward hardlimit neurons and three feedback neurons, which are used to determine the dynamic threshold. From theoretical analysis and simulation results, we found that the convergence of the proposed KWTA network, which requires Log2(N+1) iterations in average to complete a KWTA process, is independent of K, the number of the desired winners, and faster than that of the existing KWTA networks
  • Keywords
    digital simulation; feedback; neural nets; K-winning neurons; dynamic K-winners-take-all neural network; dynamic threshold; feedback neurons; feedforward hardlimit neurons; simulation results; Associative memory; Clocks; Control systems; Convergence; Councils; Neural networks; Neurons; Resonance; Self organizing feature maps; Subspace constraints;
  • fLanguage
    English
  • Journal_Title
    Systems, Man, and Cybernetics, Part B: Cybernetics, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    1083-4419
  • Type

    jour

  • DOI
    10.1109/3477.584959
  • Filename
    584959