Morphological model of self-similar multilayer neural networks

Abstract

A class of multilayer modular neural networks with self-similar structure is considered. The paper introduces the concept of morphogenesis and network regularity. Conditions for morphogenesis of the multilayer network are determined. The theorem on morphology of the weakly connected multilayer networks is proved, and invariants of system graphs for the regular self-similar networks are obtained. A rule for graph construction of the self-similar multilayer networks is proposed. It is noted that self-similar networks describe the structure of fast Fourier transform algorithms.