Abstract
A recursive orthogonal least squares (ROLS) algorithm for multi-input, multi-output systems is developed in this paper and is applied to updating the weighting matrix of a radial basis function network. An illustrative example is given, to demonstrate the effectiveness of the algorithm for eliminating the effects of ill-conditioning in the training data, in an application of neural modelling of a multi-variable chemical process. Comparisons with results from using standard least squares algorithms, in batch and recursive form, show that the ROLS algorithm can significantly improve the neural modelling accuracy. The ROLS algorithm can also be applied to a large data set with much lower requirements on computer memory than the batch OLS algorithm.
Similar content being viewed by others
References
J.B. Gomm, D.L. Yu and D. Williams, “A new model structure selection method for non-linear systems in neural modelling”, Proc. UKACC International Conference on CONTROL'96, pp. 752–757, Exeter, UK, Institution of Electrical Engineers, 1996.
J.B. Gomm, D. Williams, J.T. Evans, S.K. Doherty and P.J.G. Lisboa, “Enhancing the non-linear modelling capabilities of MLP neural networks using spread encoding”, Fuzzy Sets and Systems, Vol. 79, No. 1, pp. 113–126, 1996.
M.J. Korenberg, S.A. Billings, Y.P. Liu and P.J. McIloy, “Orthogonal parameter estimation algorithm for non-linear stochastic systems”, Int. J. Control, Vol. 48, pp. 193–210, 1988.
S. Chen, S.A. Billings, C.F.N. Cowan and P.M. Grant, “Practical identification of NARMAX models using radial basis functions”, Int. J. Control, Vol. 52, No. 6, pp. 1327–1350, 1990.
J.E. Bobrow and W. Murray, “An algorithm for RLS identification of parameters that vary quickly with time”, IEEE Trans. Automatic Control, Vol. 38, No. 2, pp. 351–354, 1993.
L. Ljung and T. Soderstrom, Theory and Practice of Recursive Identification, MIT Press: Cambridge, MA, 1983.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Yu, D., Gomm, J. & Williams, D. A Recursive Orthogonal Least Squares Algorithm for Training RBF Networks. Neural Processing Letters 5, 167–176 (1997). https://doi.org/10.1023/A:1009622226531
Issue Date:
DOI: https://doi.org/10.1023/A:1009622226531