Research map / Multi-agent systems and games / Games and equilibria
Global Convergence of Policy Gradient for Linear-Quadratic Mean-Field Control/Game in Continuous Time
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use model-based policy gradient for the linear feedback policy | Continuous-time LQR · Continuous-time LQR | cost convergence | Helped |
| Reparametrize the policy and state into conditional-mean and deviation components | Linear-quadratic mean-field control · Linear-quadratic mean-field control | cost convergence | Helped |
| Alternate best-response policy updates with mean-field state/action updates | Linear-quadratic mean-field game · Linear-quadratic mean-field game | Nash equilibrium convergence | Helped |
| Require the mean-field update operator to be a contraction under the stated regularity conditions | Linear-quadratic mean-field game · Linear-quadratic mean-field game | Nash equilibrium uniqueness: L_0 < 1 | Helped |
| Allow agents to optimize individually rather than jointly control the mean field | MFG versus MFC · MFG simulation | total cost | Hurt |
Closest papers by meaning
- Stochastic Semi-Gradient Descent for Learning Mean Field Games with Population-Aware Function Approximation (ICLR 2025 · similarity 0.91)
- Finite-Sample Convergence Bounds for Trust Region Policy Optimization in Mean Field Games (ICML 2025 · similarity 0.90)
- Major-Minor Mean Field Multi-Agent Reinforcement Learning (ICML 2024 · similarity 0.90)
- Solving Continuous Mean Field Games: Deep Reinforcement Learning for Non-Stationary Dynamics (NeurIPS 2025 · similarity 0.89)
- Learning Discrete-Time Major-Minor Mean Field Games (AAAI 2024 · similarity 0.88)
- Last Iterate Convergence in Monotone Mean Field Games (NeurIPS 2025 · similarity 0.88)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.