Exploiting the separability of linear and nonlinear parameters in radial basis function networks

Ferreira, P. M.; Ruano, Antonio

http://hdl.handle.net/10400.1/2151

Use this identifier to reference this record.

Name:	Description:	Size:	Format:
ferreira_RBF.pdf		66.89 KB	Adobe PDF	Download

Send Feedback

Authors

Ferreira, P. M.

Ruano, Antonio

Abstract(s)

In intelligent control applications, neural models and controllers are usually designed by performing an off-line training, and then adapting it on-line when placed in the operating environment. It is therefore of crucial importance to obtain a good off-line model by means of a good off-line training algorithm. In this paper a method is presented that fully exploits the linear-nonlinear structure found in Radial Basis Function networks, being additionally applicable to other feed-forward supervised neural networks. The new algorithm is compared with two known hybrid methods.

URI

http://hdl.handle.net/10400.1/2151

Citation

Ferreira, P. M.; Ruano, A. E. Exploiting the separability of linear and nonlinear parameters in radial basis function networks, Trabalho apresentado em Symposium on Adaptive Systems for Signal Processing Communications and Control, In Proceedings of the IEEE 2000 Adaptive Systems for Signal Processing, Communications, and Control Symposium (Cat. No.00EX373), Lake Louise, Alta., Canada, 2000.