A network of chaotic elements for information processing
Neural Networks
Symmetry Detection Using Global-Locally Coupled Maps
ICANN '02 Proceedings of the International Conference on Artificial Neural Networks
A new learning method for S-GCM
AI'06 Proceedings of the 19th Australian joint conference on Artificial Intelligence: advances in Artificial Intelligence
Hi-index | 0.01 |
Based on an analysis of current principal chaotic neural network models and their applications in information processing, we propose a one-dimensional, two-way coupled map network and a modified definition of an auto-associative matrix. The two-way coupled map network overcomes the weakness of the globally coupled map network and has the same abilities in pattern classification. Numerical simulation experiments have shown that the associative success rate and recall speed of our modified definition of the auto-associative matrix are an improvement them over existing methods. Moreover, the associative recall process of the network is analyzed in detail and explanations of improvement are given, basd on our theoretical analysis.