Noncooperative Dynamic Games with a Large Population of Players
Abstract
This proposal addresses a number of fundamental problems that arise in connection with non cooperative nonzero sum dynamic and differential games with asymmetric information across possibly a heterogeneous group of players (agents) and with possibly multiple layers in decision making. The focus is on stochastic games with a large population of players, and study of the relationship between such games and corresponding ones with an infinite population, particularly mean field games. Issues of paramount importance in this context, which will be addressed in the proposed three year program, are related to: the role of dynamic information acquired by the players over noisy channels with possible bandwidth limitations; leakage and adversarial intervention; robustness to unmodeled elements of the game; control by upper layers the behavior that would emerge from actions of infinitesimal (but high population of) players; the role of network topologies that govern the interactions among players through neighborhood associations; and limited sharing of structural information among the players and misalignment in players’ beliefs and perceptions.
Document Details
- Document Type
- DoD Grant Award
- Publication Date
- Jan 14, 2022
- Source ID
- FA95501910353
Entities
People
- Tamer Başar
Organizations
- Air Force Office of Scientific Research
- United States Air Force
- University of Illinois Urbana–Champaign