Associate Professor Nik Kasabov and  Mike Watts
Department of Information Science

Title:  Fuzzy neural networks, evolving fuzzy neural networks and how genetic algorithms can help

Abstract:  
The major concepts of fuzzy neural networks FuNNs and evolving fuzzy neural networks EFuNNs are presented and illustrated on several real-world problems, such as adaptive phoneme-based speech recognition,  system identification, adaptive control, intelligent adaptive agents.  EFuNNs allow for on-line, incremental, life-long learning in dynamically changing environment. Genetic algorithms are employed to optimise the parameters of FuNNs and EFuNNs. The presentation includes a demonstration of a recently developed software system. 

When:  Friday, 1.00 pm, 9 October 1998

Where:	  Castle C