In this paper relation between complex systems, information theory and the simplest models of neural networks are elucidated. Two different types of complex systems are distinguished, new complexity measure based on the graph theory defined, hierarchy of the correlation matrices introduced and connection with the correlation matrix memories and other types of neural models explained.
Reference: Duch W and Jankowski N (1994) Complex systems, Information Theory and Neural Networks. In: I Polish Conference on Neural Networks and Their Applications, Kule, Poland, 12-15.04.1994, pp. 224-230
Paper in PDF format, 181 KB
Projects on similar subject and BACK to the on-line publications of W. Duch.