Abstract
This article describes a new structure to create a RBF neural network that uses regression weights to replace the constant weights normally used. These regression weights are assumed to be functions of input variables. In this way the number of hidden units within a RBF neural network is reduced. A new type of nonlinear function is proposed: the pseudo-gaussian function. With this, the neural system gains flexibility, as the neurons possess an activation field that does not necessarily have to be symmetric with respect to the centre or to the location of the neuron in the input space. In addition to this new structure, we propose a sequential learning algorithm, which is able to adapt the structure of the network; with this, it is possible to create new hidden units and also to detect and remove inactive units. We have presented conditions to increase or decrease the number of neurons, based on the novelty of the data and on the overall behaviour of the neural system, (for example, pruning the hidden units that have lowest relevance to the neural system using Orthogonal Least Squares (OLS) and other operators), respectively. The feasibility of the evolution and learning capability of the resulting algorithm for the neural network is demonstrated by predicting time series.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Anouar, F., Badran, F., Thiria, S.: Probabilistic self-organizing maps and radial basis function networks. Neurocomputing 20, 83–96 (1998)
Benaim, M.: approximation with normalized Gaussian units. Neural Comput. 6 (1994)
Chen, S., Cowan, C.F.N., Grant, P.M.: Orthogonal Least Squares Learning Algorithm for Radial Basis Functions. IEEE Trans. Neural Networks 2, 302–309 (1999)
Cho, K.B., Wang, B.H.: Radial basis function based adaptive fuzzy systems and their applications to system identification and prediction. Fuzzy Sets and Systems 83, 325–339 (1995)
Haykin, S.: Neural Networks- A Comprehensive Foundation. IEEE Press, New York (1994)
Iokibe, T., Fujimoto, Y., Kanke, M., Suzuki, S.: Short-Term prediction of chaotic time series by local fuzzy reconstruction method. Journal of Intelligent and Fuzzy Systems 5, 3–21 (1997)
Golub, G.H., Van Loan, C.F.: Matrix Computations, 3rd edn. Johns Hopkins University Press, Baltimore (1996)
Grossberg, S.: Competitive Learning: From Interactive Activation to Adaptive Resonance. Cognitive Science 11, 23–63 (1987)
Jang, J.S.R., Sun, C.T., Mizutani, E.: Neuro-Fuzzy and soft computing. Prentice Hall, Englewood Cliffs (1997) ISBN 0-13-261066-3
Karayiannis, N.B., Weiqun Mi, G.: Growing Radial Basis Neural Networks: Merging Supervised and Unsupervised Learning with Network Growth Techniques. IEEE Transaction onNeural Networks 8(6), 1492–1506 (1997)
Kim, D., Kim, C.: Forecasting time series with genetic fuzzy predictor ensemble. IEEE Transactions on Fuzzy Systems 5(4), 523–535 (1997)
Kohonen, T.: Self-Organization and Associative Memory. Springer, New York (1988)
Lee, S., Kil, R.M.: A Gaussian potential function network with hierarchically self-organizing learning. Neural Network 2, 207–224 (1991)
Lee, S.-H., Kim, I.: Time series analysis using fuzzy learning. In: Proc. Int. Conf. Neural Inform. Processing, Seoul, Korea, October 1994, vol. 6, pp. 1577–1582 (1994)
Moody, J.E., Darke, C.J.: Fast learning in networks of locally-tuned processing units. Neural Computa. 1, 281–294 (1989)
Musavi, M.T., Ahmed, W., Chan, K.H., Faris, K.B., Hummels, D.M.: On the training of radial basis function classifier. Neural Networks 5, 595–603 (1992)
Nowlan, S.: Maximum likelihood competitive learning. In: Proc. Neural Inform. Process. Systems, pp. 574–582 (1990)
Platt, J.: A resource allocating network for function interpolation. Neural Computa. 3, 213–225 (1991)
Rojas, I., Anguita, M., Ros, E., Pomares, H., Valenzuela, O., Prieto, A.: What are the main factors involved in the design of a Radial Basis Function Network? In: 6th European Symposium on Artificial Neural Network, ESANN 1998, April 22-24, pp. 1–6 (1998)
Sherstinsky, A., Picard, R.W.: On the efficiency of the orthogonal least squares training method for radial basis function networks. IEEE Trans. Neural Networks 7(1), 195–200 (1996)
Wang, L.X., Mendel, J.M.: Generating fuzzy rules by learning from examples. IEEE Trans. On Syst. Man and Cyber. 22(6), 414–1427 (1992)
Whitehead, B.A., Choate, T.D.: Cooperative-Competitive Genetic Evolution of Radial Basis Function Centers and Widths for Time Series Prediction. IEEE Transaction on Neural Networks 7(4), 869–880 (1996)
Yingwei, L., Sundarajan, N., Saratchandran, P.: Performance Evaluation of a Sequential Minimal Radial Basis Function (RBF) Neural Network Learning Algorithm. IEEE Transactions on Neural Networks 9(2), 308–318 (1998)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2003 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Valenzuela, O., Rojas, I., Rojas, F. (2003). Sequential Learning Algorithm of Neural Networks Systems for Time Series. In: Günter, A., Kruse, R., Neumann, B. (eds) KI 2003: Advances in Artificial Intelligence. KI 2003. Lecture Notes in Computer Science(), vol 2821. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-39451-8_24
Download citation
DOI: https://doi.org/10.1007/978-3-540-39451-8_24
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-20059-8
Online ISBN: 978-3-540-39451-8
eBook Packages: Springer Book Archive