Mean Field Game Theory for Agents with Individual-State Partial Observations

TL;DR

This paper develops a mean field game framework for large populations where each agent has only partial observations of its state, using nonlinear filtering and the separation principle to establish approximate Nash equilibria.

Contribution

It introduces a novel analysis of mean field games with partial state observations, incorporating the agent's control-dependent filtering process.

Findings

Existence of an $oldsymbol{ ext{ extit{epsilon}}}$-Nash equilibrium with partial observations.

Control actions depend on the filtered estimate of the agent's state and the mean field.

Comparison with previous models shows the control-dependent state process in this work.

Abstract

Subject to reasonable conditions, in large population stochastic dynamics games, where the agents are coupled by the system's mean field (i.e. the state distribution of the generic agent) through their nonlinear dynamics and their nonlinear cost functions, it can be shown that a best response control action for each agent exists which (i) depends only upon the individual agent's state observations and the mean field, and (ii) achieves a $\epsilon$-Nash equilibrium for the system. In this work we formulate a class of problems where each agent has only partial observations on its individual state. We employ nonlinear filtering theory and the Separation Principle in order to analyze the game in the asymptotically infinite population limit. The main result is that the $\epsilon$-Nash equilibrium property holds where the best response control action of each agent depends upon the conditional density of its own state generated by a nonlinear filter, together with the system's mean field. Finally, comparing this MFG problem with state estimation to that found in the literature with a major agent whose partially observed state process is independent of the control action of any individual agent, it is seen that, in contrast, the partially observed state process of any agent in this work depends upon that agent's control action.