Aug 16, 2020 · In this paper, we study the policy gradient method for the linear-quadratic mean-field control and game, where we assume each agent has ...

To present its formal definition, we define Λ1(µ) as the optimal policy in Π given the mean- field µ, and define Λ2(µ, π) as the mean-field state/action.

Aug 16, 2020 · To present its formal definition, we define Λ1(µ) as the optimal policy in Π given the mean-field state µ, and define Λ2(µ, π) as the mean-field ...

Aug 16, 2020 · Therefore, it has motivated new research directions for mean-field control (MFC) and mean-field game (MFG). In this paper, we study the policy ...

Jul 19, 2021 · Global Convergence of Policy Gradient for Linear-Quadratic Mean-Field Control/Game in Continuous Time · Speakers · Organizer · About ICML 2021.

Supplementary Material: Global Convergence of Policy Gradient for. Linear-Quadratic Mean-Field Control/Game in Continuous Time. Weichen Wang. ∗. , Jiequn Han.

This work proves rigorously the convergence of exact and model-free policy gradient methods in a mean-field linear-quadratic setting and provides graphical ...

We propose geometry-aware gradient descents for the mean and covariance of the policy using the Fisher geometry and the Bures–Wasserstein geometry, respectively ...

May 3, 2023 · 148. Wang W, Han J, Yang Z, Wang Z 2021. Global convergence of policy gradient for linear-quadratic mean-field control/game in continuous time.

We study discrete-time linear-quadratic MARL under mean-field settings with exchangeable finite n ... Mean field games. Japanese journal of mathematics, 2.