• DocumentCode
    1242501
  • Title

    Learning in multilayered networks used as autoassociators

  • Author

    Bianchini, M. ; Frasconi, P. ; Gori, M.

  • Author_Institution
    Dipartimento de Sistemi e Inf., Firenze Univ., Italy
  • Volume
    6
  • Issue
    2
  • fYear
    1995
  • fDate
    3/1/1995 12:00:00 AM
  • Firstpage
    512
  • Lastpage
    515
  • Abstract
    Gradient descent learning algorithms may get stuck in local minima, thus making the learning suboptimal. In this paper, we focus attention on multilayered networks used as autoassociators and show some relationships with classical linear autoassociators. In addition, by using the theoretical framework of our previous research, we derive a condition which is met at the end of the learning process and show that this condition has a very intriguing geometrical meaning in the pattern space
  • Keywords
    content-addressable storage; learning (artificial intelligence); multilayer perceptrons; autoassociators; geometrical meaning; gradient descent learning algorithms; multilayered networks; pattern space; Backpropagation; Convergence; Costs; Intelligent networks; Linearity; Neurons; Pattern analysis; Rough surfaces; Shape; Surface roughness;
  • fLanguage
    English
  • Journal_Title
    Neural Networks, IEEE Transactions on
  • Publisher
    ieee
  • ISSN
    1045-9227
  • Type

    jour

  • DOI
    10.1109/72.363492
  • Filename
    363492