Elisa Calzola,Elisabetta Carlini,Francisco J. Silva
Abstract:In this paper we propose a high-order numerical scheme for time-dependent mean field games systems. The scheme, which is built by combining Lagrange-Galerkin and semi-Lagrangian techniques, is consistent and stable for large time steps compared with the space steps. We provide a convergence analysis for the exactly integrated Lagrange-Galerkin scheme applied to the Fokker-Planck equation, and we propose an implementable version with inexact integration. Finally, we validate the convergence rate of the proposed scheme through the numerical approximation of two mean field games systems.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is to develop a high - order numerical scheme for solving the time - dependent Mean Field Games (MFGs) system. Specifically, the author proposes a high - order numerical scheme that combines Lagrange - Galerkin and semi - Lagrangian techniques to improve the numerical approximation accuracy of the MFG system.
### Problem Background
Mean Field Games (MFGs) are a class of mathematical models introduced by Lasry - Lions and Huang - Caines - Malhamé respectively, which are used to describe Nash equilibria in stochastic differential games with an infinite number of indistinguishable players. In some specific cases, these equilibria can be described by a system of partial differential equations (PDEs), which is a coupling of the Hamilton - Jacobi - Bellman (HJB) equation with terminal conditions and the Fokker - Planck (FP) equation with initial conditions.
### Deficiencies in Existing Research
Although a great deal of research has been carried out on the numerical approximation of MFG systems in recent years, most of the work has focused on low - order or medium - order methods. At present, only a few studies involve high - order numerical schemes, such as the finite difference method and the space - time finite element method. Therefore, developing an efficient, stable and accurate high - order numerical method is of great significance for further promoting the research of MFGs.
### Main Contributions of the Paper
1. **Proposing a new high - order numerical format**: This format can handle second - order MFG systems and has a convergence order greater than second - order.
2. **Combining Lagrange - Galerkin and semi - Lagrangian techniques**: By using the Crank - Nicolson method to discretize the time variable of the FP equation and symmetric odd - order Lagrange basis functions to discretize the spatial variable, the explicitness, conservation, consistency and stability of the format are ensured.
3. **Providing a detailed convergence analysis**: It is proved that the format is stable and convergent under large time steps, and its high - order accuracy is verified by numerical experiments.
4. **Application examples**: The effectiveness of the proposed method is verified through two specific MFG systems (one with non - local coupling and an analytical solution, and the other with local coupling and no explicit solution).
### Mathematical Formula Representation
- FP equation:
\[
\partial_t m-\frac{\sigma^2}{2}\Delta m+\text{div}(b m) = 0\quad\text{in }(0, T)\times\mathbb{R}^d,
\]
where \(m(0,\cdot)=m_0\).
- HJB equation:
\[
-\partial_t v-\frac{\sigma^2}{2}\Delta v + H(x,\nabla v)=F(x,m(t))\quad\text{in }[0, T)\times\mathbb{R}^d,
\]
where \(v(T,\cdot)=G(\cdot,m(T))\).
### Conclusion
By combining Lagrange - Galerkin and semi - Lagrangian techniques, this paper proposes a high - order numerical scheme suitable for MFG systems, and verifies its effectiveness and superiority through theoretical analysis and numerical experiments. This provides a powerful tool for more complex and efficient MFG numerical simulations in the future.