Control of hierarchical mean field type games

Voir Plus Voir Moins

Afficher le Résumé

This thesis focuses on solving hierarchical leadership mean-field-type games with jump diffusion state equations and two different performing functionals using two different methods, concluding with a numerical analysis of the impact of the hierarchy on the final solution. In Paper A, we provide a solution for a hierarchical mean-field type optimal control problem with polynomial performing functional in three forms using the dynamic programming principle method. The first form is a one-level mean- field type game when the player acts simultaneously without information differences. The second form is a Stackelberg mean field type game, where the players are divided into two groups in two levels, one acting first, then the other reacting.