Use of some sensitivity criteria for choosing networks with good generalization ability
Purchase on Springer.com
$39.95 / €34.95 / £29.95*
Rent the article at a discountRent now
* Final gross prices may vary according to local VAT.
In most applications of the multilayer perceptron (MLP) the main objective is to maximize the generalization ability of the network. We show that this ability is related to the sensitivity of the output of the MLP to small input changes. Several criteria have been proposed for the evaluation of the sensitivity. We propose a new index and present a way for improving these sensitivity criteria. Some numerical experiments allow a first comparison of the efficiencies of these criteria.
- G. Cybenko. Approximations by superpositions of a sigmoidal function,Mathematics of Control, Signals and Systems, vol. 2, pp. 303–314, 1989.
- J.H. Friedman. An overview of predictive learning and function approximation, in V. Cherkassky, J.H. Friedman, H. Wechsler eds.,From Statistics to Neural Networks, Theory and Pattern Recognition Applications, Springer-Verlag, pp. 1–61, 1993.
- A. S. Weigend, B. A. Huberman and D. E. Rumelhart. Predicting sunspots and exchange rates with connectionist networks, in M. Casdagli and S. Eubank eds.,Nonlinear Modeling and Forecasting, Addison-Wesley, pp. 395–432, 1992.
- L. Fu and T. Chen. Sensitivity analysis for input vector in multilayer feedforward neural networks,IEEE Internatioal Conférence on Neural Networks, San Francisco, California, pp. 215–218, 1993.
- K. Matsuoka. An approach to generalization problem in back-propagation learning,Proc. INNC'90, Paris, pp. 765–768, 1990.
- C. M. Bishop. Curvature-driven smoothing in backpropagation neural networks,Proc. INNC'90, Paris, pp. 749–752, 1990.
- A. R. Webb. Functional approximation by feedforward networks: a least-squares approach to generalization,IEEE Trans. on Neural Networks, vol. 5, no. 3, pp 363–371, 1994.
- Use of some sensitivity criteria for choosing networks with good generalization ability
Neural Processing Letters
Volume 2, Issue 6 , pp 1-4
- Cover Date
- Print ISSN
- Online ISSN
- Kluwer Academic Publishers
- Additional Links
- Industry Sectors
- Author Affiliations
- 1. UMR 9964, Equipe de Biologie Quantitative, Université Paul Sabotier, 118 Route de Narbonne, F-31062, Toulouse Cedes, France
- 2. Département d'Etudes et de Recherches en Informatique, ONERA-CERT, 2, av. Edouard Belin, F-31055, Toulouse Cedex, France