Abstract
In this work we present new approaches for the optimal identification of nonlinear systems. We optimize different parameters of feedforward neural networks and of the learning schedule backpropagation by the use of global search methods like genetic algorithms and simulated annealing. We achieve a global increment of their learning capability thereby enlarging the generalization capability and reducing the amount of learning speed.
The result is a more reliable and robust model for nonlinear systems.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
Literature
Burrows, T.L.; Niranjan, M.: The use of feed-forward and recurrent neural networks for system identification. Technical Report TR-158, University of Cambridge, U.K., 1993.
Goldberg, D.: Genetic Algorithms in Search, Optimization & Machine Learning, Addison-Wesley, 1989.
Hornik, K.: Multilayer feedforward networks are universal approximators. Neural Networks, Vol. 2, pp. 359–366, 1989.
Kirkpatrick, S.; Gelatt, C.; Vecchi, M.: Optimization by Simulated Annealing. Science 220, pp. 671–680, 1983.
Ljung, L., System Identification. Prentice-Hall, 1987.
Narendra, K.S.; Parthasarathy, K.: Identification and control of dynamic systems using neural networks. IEEE Trans. on Neural Networks, Vol 1, N∘ 1, pp. 4–27, 1990.
Rumelhart, D.E., Hinton, G.E., Williams R.J. Learning internal representations by error propagation. In Parallel Distributing Processing, Vol. 1, pp. 318–362, MIT Press, 1986.
Sinne, S.: M.Sc. Thesis (in Progress), Dept. Computer Science I, University of Dortmund, 1994.
Vergara, V., Moraga, C., Computational Intelligence for the Identification of Dynamical Systems, in Proceedings of the 3dr Int. Conf. on Fuzzy Logic, Neural Nets and Soft Computing, Iizuka, Japan, 1994.
Wassermann, P.D.: Neural computing: theory and praxis. Van Nostrand Reinhold, 1989.
Zwicker, E.: Simulation und Analyse dynamischer Systeme in den Wirtschaft-und Sozialwissenschaften, Berlin, 1981.
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1995 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Vergara, V., Sinne, S., Moraga, C. (1995). Optimal identification using feed-forward neural networks. In: Mira, J., Sandoval, F. (eds) From Natural to Artificial Neural Computation. IWANN 1995. Lecture Notes in Computer Science, vol 930. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-59497-3_284
Download citation
DOI: https://doi.org/10.1007/3-540-59497-3_284
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-59497-0
Online ISBN: 978-3-540-49288-7
eBook Packages: Springer Book Archive