Title :
FAN: learning by means of free associative neurons
Author :
Souza, J.A. ; Dandolini, G.A. ; Pacheco, R.C.S. ; Martins, Alfredo
Abstract :
We introduce a learning method: free associative neurons (FAN). FAN is composed of independent units with autonomous learning capability. The learning power of FAN is based on the association of its units and on the use of granularity for representing information. As a method for representing complex environments, FAN can be used in pattern recognition, classification and diagnosis. The paper is focused on the principles governing FAN. We also discuss some results compared to well-known artificial neural network algorithms
Keywords :
fuzzy set theory; learning (artificial intelligence); neural nets; autonomous learning capability; classification; complex environments; diagnosis; free associative neurons; granularity; information representation; learning method; learning power; pattern recognition; Chaos; Competitive intelligence; Computational intelligence; Humans; Learning systems; Neural networks; Neurons; Pattern recognition; Production engineering; Uncertainty;
Conference_Titel :
Neural Networks Proceedings, 1998. IEEE World Congress on Computational Intelligence. The 1998 IEEE International Joint Conference on
Conference_Location :
Anchorage, AK
Print_ISBN :
0-7803-4859-1
DOI :
10.1109/IJCNN.1998.682304
