Integration of chaotic behavior modules into a flow-typed two-body neural network model
Akira Sano, Susumu Kunifuji · Electronics and Communications in Japan (Part III Fundamental Electronic Science) · 1998
Features of information processing in the brain include distributed functional structure, as in the areas and columns, as well as integrated processing among those parts. This paper presents a model, where one-way coupling is provided between two-body modules, as a basic structure for the neural network model composed of multiple modules. The effective role of chaos is demonstrated as the integration mechanism of the neural network model. The relation between the coupling density between modules and the pattern recall behavior in the two-body one-way coupling model is analyzed. It is verified that there exists an area, called the A area, of the intermodule coupling parameters, where the patterns can be retained in both modules. © 1998 Scripta Technica. Electron Comm Jpn Pt 3, 81(5): 41–50, 1998