Creating Self-organization Maps by Cooperative Information Control

This paper proposes a novel information theoretic approach to self-organization called cooperative information control. The method aims to mediate between competition and cooperation among neurons by controlling information content in neurons. Competition is realized by maximizing information content in neurons. In the process of information maximization, only a small number of neurons win the competition, while all the others are inactive. Cooperation is implemented by having neurons behave similarly to their neighbors. These two processes are uni ed and controlled in the framework of cooperative information control. We applied the new method to a political analysis. In the analysis, experimental results con rmed that competition and cooperation are exibly controlled. In addition, controlled processes can yield a number of di erent neuron ring patterns, which can be used to detect macro as well as micro features in input patterns.