Abstract: The paper formulates and numerically studies the direct and inverse problems for the mean field game model, which describes the process of information dissemination in online social networks taking into account external influences. The mean field game model is reduced to a joint solution of initial-boundary value problems for the Kolmogorov-Fokker-Planck (KFP) and HamiltonJacobi-Bellman (HJB) equations, as well as the Nash optimality condition. Based on the Sobol global sensitivity analysis, the sensitivity of the control parameters and the initial condition of the KFP problem to the function of measuring the user density at a fixed time from the moment of information dissemination is shown. Analgorithm for numerically solving inverse problems for the KFP and HJB equations that differ in the control function is constructed based on Bayesian optimization. It is shown that the reconstructed initial user density of the network describes the synthetic data of the inverse problem with greater accuracy in the case of fixed control, which is optimal for the exact solution of the inverse problem.

Cite: Звонарева Т.А. , Криворотько О.И.
Моделирование распространения информации в рамках принципа среднего поля
Сибирские электронные математические известия (Siberian Electronic Mathematical Reports). 2025. Т.22. №1. С.307-325. DOI: 10.33048/semi.2025.22.021 WOS

Dates:

Submitted:	Oct 27, 2024
Published online:	Dec 31, 2024
Published print:	Apr 14, 2025

Identifiers:

Web of science:

WOS:001473623500018

Citing: Пока нет цитирований

Altmetrics: