Hi, are there any MARL examples that account for hundreds of agents / population based learning ?
Not sure of an particular examples for RLlib, but I do believe that you can scale Multiagent RL to 100s of agents if they all use PPO.
In addition, there is prior research on scaling MARL here: [1903.04527] Multi-Agent Deep Reinforcement Learning for Large-scale Traffic Signal Control
1 Like