Title :
Using classifier systems to design neural nets
Author :
Nikolopoulos, Chris
Author_Institution :
Dept. of Comput. Sci. & Artificial Intelligence, Malta Univ., Valetta, Malta
Abstract :
A new neural network training algorithm, based on the use of genetic algorithms and classifier systems, is presented. The algorithm not only calculates the weights of the network links, but also automates the traditionally ad hoc, trial and error method for finding the most appropriate architecture of the neural net
Keywords :
Accuracy; Convergence; Fires; Genetics; Liver diseases; Neural networks; Neurons; Pattern matching; Testing;
Conference_Titel :
Tools with Artificial Intelligence, 1994. Proceedings., Sixth International Conference on
Conference_Location :
New Orleans, LA
Print_ISBN :
0-8186-6785-0
DOI :
10.1109/TAI.1994.346395
