Back to search
2111.11972

A semi-discrete approximation for first-order stationary mean field games

Renato Iturriaga, Kaizhi Wang

correctmedium confidence
Category
Not specified
Journal tier
Specialist/Solid
Processed
Sep 28, 2025, 12:56 AM

Audit review

The paper proves (i) existence of m_τ via Kakutani’s fixed point on τ-holonomic minimizers and (ii) convergence (along subsequences) of (u_{τ,m_τ}, m_τ) to a stationary MFG solution, by establishing tightness, uniform Lipschitz bounds, convergence to the Hamilton–Jacobi equation, and passing τ-holonomy to closed measures to obtain the continuity equation. These steps are explicit in Theorem 1 and Propositions 6–11 (tightness, Kakutani map upper semicontinuity, viscosity limit, closed-limit/Mather measure, and divergence-free flow) . The candidate solution establishes the same two claims with a slightly different route: standard viscosity stability (Crandall–Ishii–Lions), a calibration/Fenchel–Young argument to identify v = ∂_p H on the limit measure’s support, and a direct distributional pass-to-the-limit of the τ-holonomy constraint. Aside from minor imprecisions (e.g., streamlining the proof that the limit measure is minimizing for L_{m_0}), the model matches the paper’s result and logic.

Referee report (LaTeX)

\textbf{Recommendation:} minor revisions

\textbf{Journal Tier:} specialist/solid

\textbf{Justification:}

The paper presents a clear semi-discrete (in time) approximation scheme for first-order stationary mean field games on the torus and proves convergence to weak KAM/Mather-type stationary solutions. The strategy is carefully adapted from discrete Aubry–Mather/weak KAM theory and extended to include the MFG coupling through assumptions (F1)–(F3). Existence of a fixed point for the marginal of minimizing τ-holonomic measures is obtained by Kakutani’s theorem, and tightness plus operator estimates enable the passage to the HJ limit; the continuity equation is derived by identifying the limit measure as a Mather measure and invoking invariance arguments. The contribution is technically solid and well aligned with existing weak KAM machinery; some steps could be streamlined or more explicitly connected to standard references.