• DocumentCode
    1333588
  • Title

    Gradient evaluation for neural-networks-based electromagnetic optimization procedures

  • Author

    Antonini, G. ; Orlandi, A.

  • Author_Institution
    Dept. of Electr. Eng., Univ. of l´´Aquila, Italy
  • Volume
    48
  • Issue
    5
  • fYear
    2000
  • fDate
    5/1/2000 12:00:00 AM
  • Firstpage
    874
  • Lastpage
    876
  • Abstract
    This paper extends the use of a neural network (NN) approximating a function, to the evaluation of the gradient of the same function. This is done without any extra training of the network. The evaluation of the function´s gradient is used in NN-based optimization procedures in order to speed up the convergence and to maintain the overall accuracy
  • Keywords
    convergence; electrical engineering computing; electromagnetism; neural nets; optimisation; artificial neural networks; convergence; electromagnetic optimization procedures; function gradient evaluation; neural-network-based EM optimization procedures; Artificial neural networks; Conductors; Coplanar waveguides; Dielectric constant; Dielectric thin films; Gallium arsenide; Impedance; Neural networks; Optimization methods; Power transmission lines;
  • fLanguage
    English
  • Journal_Title
    Microwave Theory and Techniques, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    0018-9480
  • Type

    jour

  • DOI
    10.1109/22.841892
  • Filename
    841892