• DocumentCode
    328902
  • Title

    "Dynamic perceptron". An outlook of computational dynamics

  • Author

    Perrone, Antonio L. ; Basti, Gianfranco

  • Author_Institution
    Dept. of Phys., Istituto Nazionale di Fisica Nucl., Rome, Italy
  • Volume
    2
  • fYear
    1993
  • fDate
    25-29 Oct. 1993
  • Firstpage
    1377
  • Abstract
    With respect to Rosenblatt linear perceptron, a classical limitation theorem demonstrated by Minsky-Papert is discussed. This theorem, "ΨOne-in-a-box", ultimately concern the intrinsic "mathematical" limitations of parallel calculations in pattern recognition problems. We demonstrate a possible solution of this limitation problem by substituting the static definition of characteristic functions and of their domains in the "geometrical" perceptron, with their dynamic definition. This dynamics consists in the mutual redefinition of the characteristic function and of its domain depending on the matching with the input. We show an application of this "dynamic" perceptron scheme in particle tracks recognition in high energy physics. This algorithm has actually been used for real time automatic triggering of ADONE e+e- storage ring (Frascati, Rome) to evaluate the neutron time-like electromagnetic form factor in the context of "Fenice" collaboration by Italian Institute of Nuclear Physics (INFN).
  • Keywords
    high energy physics instrumentation computing; nuclear physics; parallel processing; particle tracks; pattern recognition; perceptrons; ADONE storage ring; Rosenblatt linear perceptron; dynamic perceptron; dynamics; high energy physics; parallel calculations; particle tracks recognition; pattern recognition; Concurrent computing; Impedance matching; Neural networks; Nuclear physics; Parallel architectures; Particle tracking; Pattern recognition; Physics computing; Storage rings; Topology;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Neural Networks, 1993. IJCNN '93-Nagoya. Proceedings of 1993 International Joint Conference on
  • Print_ISBN
    0-7803-1421-2
  • Type

    conf

  • DOI
    10.1109/IJCNN.1993.716800
  • Filename
    716800