• DocumentCode
    296021
  • Title

    MLP classifiers: overtraining and solutions

  • Author

    Chi, Zheru

  • Author_Institution
    Dept. of Electron. Eng., Hong Kong Polytech., Hung Hom, Hong Kong
  • Volume
    5
  • fYear
    1995
  • fDate
    Nov/Dec 1995
  • Firstpage
    2821
  • Abstract
    Training a multi-layer perceptron (MLP) classifier is difficult to control and as a result its performance on unseen patterns is unpredicted. Overtraining is one of many problems in training an MLP classifier. In this paper, the author first discusses the overtraining problem based on an artificial two-input two-category classification problem. The author then suggests five solutions to the overtraining problem, which are supported by experimental results
  • Keywords
    learning (artificial intelligence); multilayer perceptrons; pattern classification; multi-layer perceptron classifier; overtraining; two-input two-category classification problem; Degradation; Electronic mail; Multilayer perceptrons; Pattern recognition; Testing;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Neural Networks, 1995. Proceedings., IEEE International Conference on
  • Conference_Location
    Perth, WA
  • Print_ISBN
    0-7803-2768-3
  • Type

    conf

  • DOI
    10.1109/ICNN.1995.488180
  • Filename
    488180