The Analysis of Artificial Neural Network Structure Recovery Possibilities Based on the Theory of Graphs

In the paper, the problem of the possibility of recovering the unknown structure of artificial neural networks (ANNs) using the theory of graphs is investigated. The key ANN concepts, their typical architectures, and differences are considered. The application of the theory of the graph tool for solving the problem of detecting an ANN structure is substantiated, and examples of comparing different ANN architectures and graph types are presented. It is proposed to use the methods of the spectral graph theory and the graph signal processing as tools for analyzing the ANN structure.