Supervised training algorithms for B-spline neural networks and fuzzy systems

Ruano, Antonio; Cabrita, Cristiano Lourenço; Oliveira, J. V.; Tikk, D.; Kóczy, László T.

http://hdl.handle.net/10400.1/2210

Use this identifier to reference this record.

Name:	Description:	Size:	Format:
IFSA.pdf		508.31 KB	Adobe PDF	Download

Send Feedback

Authors

Ruano, Antonio

Cabrita, Cristiano Lourenço

Oliveira, J. V.

Tikk, D.

Kóczy, László T.

Abstract(s)

Complete supervised training algorithms for B-spline neural networks and fuzzy rule-based systems are discussed. By introducing the relationships between B-spline neural networks and Mamdani (satisfying certain assumptions) fuzzy model, training algorithms developed initially for neural networks can be adapted to fuzzy systems. The standard training criterion is reformulated, by separating the linear and nonlinear parameters. By employing this reformulated criterion with the Levenberg-Marquardt algorithm, a new training method, offering a fast rate of convergence is obtained. It is also shown that the standard Error-Back Propagation algorithm, the most common training method for this class of systems, exhibits a very poor performance.

Keywords

Training algorithms Neural networks B-splines Fuzzy systems Llinear and nonlinear separability

URI

http://hdl.handle.net/10400.1/2210

Citation

Ruano, A. E.; Cabrita, C.; Oliveira, J. V.; Tikk, D.; Koczy, L. T. Supervised training algorithms for B-spline neural networks and fuzzy systems, Trabalho apresentado em Joint 9th IFSA World Congress and 20th NAFIPS International Conference, In Proceedings Joint 9th IFSA World Congress and 20th NAFIPS International Conference (Cat. No. 01TH8569), Vancouver, BC, Canada, 2001.