Synthesis of Artificial Neural Network Training Algorithm with Interval Target Tuning in the Implementation of the Method for Constructing a Decentralized Hierarchical Network of Modular Type

The paper describes the application of artificial neural network training algorithm with interval target tuning in the implementation of the method for constructing a decentralized hierarchical network of modular type. The main concept of combining these two methods is devoted to the idea of improving the solution of the problem of constructing a model of difficult formalized systems. The proposed hierarchical structure of the network will allow solving a complex initial problem asynchronously in time and in parallel in space, involving distributed computing resources and geographically dispersed teams of researchers.