Multi-Agent Reinforcement learning

I have a game similar to poker (but different) coded up in Matlab where the game is one Matlab function and each player's strategy is a different function I've written. I'd like to replace two of the players with reinforcement learning AI agents so the AI plays itself and hopefully optimizes. Is it possible to do this in Matlab or do I have to use Simulink? I've never used the reiinforcement learning tools in Matlab so any pointers are appreciated.

Answer 1

[Emmanouil Tzorakoleftherakis]

As of R2023b, you can do multi-agent reinforcement learning using MATLAB environments. Please take a look at this example and R2023b release notes.

[Kuanishbay Sadatdiynov]

Hello Emmanouil,

Is there a way to implement multi-agent RL by creating a custom environment from a class template rather than using reset and step functions? If yes, could you share the example, please? Thanks.

[Emmanouil Tzorakoleftherakis]

We are working on this. In the meantime, you can use the following environment as a template:

edit rl.env.pusher.PusherEnvironment

[Kuanishbay Sadatdiynov]

Thank you, let me try.