Linear Quadratic Extended Mean Field Games and Control Problems

Alain Bensoussan,Bohan Li,Sheung Chi Phillip Yam

2023-11-09

Abstract:We provide a thorough study of a general class of linear-quadratic extended mean field games and control problems in any dimensions where the mean field terms are allowed to be unbounded and there are also presence of cross terms in the objective functionals. Our investigation focuses on the unique existence of equilibrium strategies for the extended mean field problems by employing the stochastic maximum principle approach and the appropriate fixed point argument. We provide two distinct proofs, accompanied by two sufficient conditions, that establish the unique existence of the equilibrium strategy over a global time horizon. Both conditions emphasize the importance of sufficiently small coefficients of sensitivity for the cross term, of state and control, and mean field term. To determine the required magnitude of these coefficients, we utilize the singular values of appropriate matrices and Weyl's inequalities. The present proposed theory is consistent with the classical one, namely, our theoretical framework encompasses classical linear-quadratic stochastic control problems as particular cases. Additionally, we establish sufficient conditions for the unique existence of solutions to a particular class of non-symmetric Riccati equations, and we illustrate a counterexample to the existence of equilibrium strategies. Furthermore, we also apply the stochastic maximum principle approach to examine linear-quadratic extended mean field type stochastic control problems. Finally, we conduct a comparative analysis between our method and the alternative master equation approach, specifically addressing the efficacy of the present proposed approach in solving common practical problems, for which the explicit forms of the equilibrium strategies can be obtained directly, even over any global time horizon.

Optimization and Control

What problem does this paper attempt to address?

The problem that this paper attempts to solve is about the existence of the globally unique solution in Linear Quadratic Extended Mean Field Games (LQ - EMFG) and control problems. Specifically, the paper focuses on how to prove the unique existence of the equilibrium strategy in the extended mean - field problem through the Stochastic Maximum Principle (SMP) and appropriate fixed - point arguments when the objective function contains cross - terms.

The following is the main problem description of the paper:

1. **Research Background**:
- The paper explores linear quadratic extended mean - field games and control problems under multi - dimensional state processes.
- In these models, the mean - field terms are allowed to be unbounded, and there may be cross - terms in the objective function, which increases the complexity of the problem.

2. **Core Problems**:
- How to ensure the unique existence of the equilibrium strategy in the extended mean - field game problem?
- In particular, how to find the unique solution that satisfies the conditions within the global time horizon?

3. **Main Methods**:
- Use the Stochastic Maximum Principle (SMP) to derive a forward - backward stochastic differential equations system (FBSDEs).
- Provide two sufficient conditions to prove the existence of the globally unique solution of FBSDEs:
1. The first proof method based on the small sensitivity coefficient assumption.
2. The second proof method based on the homotopy method.

4. **Core Content of Mathematical Formulas**:
- The form of the objective function is:
$$
J(v)=\mathbb{E}\left[\frac{1}{2}\int_{0}^{T}x_{t}^{\top}Q_{t}x_{t}+v_{t}^{\top}P_{t}v_{t}+(x_{t}-S_{t}\mathbb{E}[\hat{x}_{t}])^{\top}\bar{Q}_{t}(x_{t}-S_{t}\mathbb{E}[\hat{x}_{t}])+\cdots\right]
$$
which involves the cross - actions among state variables, control variables and mean - field terms.

- The form of the dynamic system is:
$$
dx_{t}=\left(A_{t}x_{t}+B_{t}v_{t}+\bar{A}_{t}\mathbb{E}[\hat{x}_{t}]+\bar{B}_{t}\mathbb{E}[\hat{v}_{t}]\right)dt+\sigma_{t}dW_{t}
$$

5. **Theoretical Contributions**:
- Two different proof methods are proposed, which are based on the small sensitivity coefficient assumption and the homotopy method respectively.
- The Weyl inequality is introduced to analyze the positive - definiteness conditions in the objective function, ensuring the existence and uniqueness of the solution.
- The research on the asymmetric Riccati equation is supplemented and expanded.

In summary, the core problem of the paper is to prove the unique existence of the equilibrium strategy in linear quadratic extended mean - field games and control problems through strict mathematical methods (such as the Stochastic Maximum Principle and fixed - point theory), especially when the objective function contains cross - terms.