How to use multi-agent TD3?

RLlib
1

https://docs.ray.io/en/master/rllib-algorithms.html#available-algorithms-overview
Here’s its listed that TD3 has an option for multi-agent. Where in the repository can I access or use multi-agent TD3?

2

Hi @dstnluong,

Have a look at this multiagent example: ray/multi_agent_cartpole.py at master · ray-project/ray · GitHub

You will need to make two changes.

The first is to switch from cartpole to pendulum for the environment since TD3 only supports continuous action spaces. This involves changing lines: 19,58,85

The second is to change from PPO to TD3 on line 104.

1 Like