# Coarse correlated equilibria for continuous time mean field games in open loop strategies

arXiv: https://arxiv.org/abs/2303.16728 · HTML: https://mlsanity.com/paper/2303.16728/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Mar 2023.

## Built on by (most cited first, 1 of 1)

- [The Learning Approach to Games](https://mlsanity.com/paper/2503.00227/) (2025)

## Abstract (arXiv)

In the framework of continuous time symmetric stochastic differential games in open loop strategies, we introduce a generalization of mean field game solution, called coarse correlated solution. This can be seen as the analogue of a coarse correlated equilibrium in the N-player game, where a moderator randomly generates a strategy profile and asks the players to pre-commit to such strategies before disclosing them privately to each one of them; such a profile is a coarse correlated equilibrium if no player has an incentive to unilaterally deviate. We justify our definition by showing that a coarse correlated solution for the mean field game induces a sequence of approximate coarse correlated equilibria with vanishing error for the underlying N-player games. Existence of coarse correlated solutions for the mean field game is proved by means of a minimax theorem. An example with explicit solutions is discussed as well.