RUS  ENG
Full version
JOURNALS // Zhurnal Vychislitel'noi Matematiki i Matematicheskoi Fiziki // Archive

Zh. Vychisl. Mat. Mat. Fiz., 2025 Volume 65, Number 4, Pages 574–589 (Mi zvmmf11960)

Computer science

Dynamic self-organization in neural networks systems

S. D. Glyzin, A. Yu. Kolesov, D. D. Fedulov

Center of Integrable Systems, Demidov Yaroslavl State University, 150003, Yaroslavl, Russia

Abstract: The introduced concept of dynamic self-organization consists in the following. Suppose that there is a set of free (non-interacting) neurons, each of which is at rest or not capable of vibrational electrical activity at all. Then, being connected in a certain way in a network, these neurons can begin to generate electrical impulses. The feasibility of this phenomenon is illustrated by the example of one mathematical model, which is a certain nonlinear boundary value problem of hyperbolic type. A combination of analytical and numerical methods is used to study the attractors of the boundary value problem under consideration.

Key words: dynamic self-organization, neural network, quasinormal form, invariant torus, asymptotics, stability, buffering.

UDC: 519.633

Received: 27.10.2024
Revised: 19.12.2024
Accepted: 04.02.2025

DOI: 10.31857/S0044466925040124


 English version:
Computational Mathematics and Mathematical Physics, 2025, 65:4, 917–934

Bibliographic databases:


© Steklov Math. Inst. of RAS, 2025