search

HumanCompatibleAI / overcooked_ai

A benchmark environment for fully cooperative human-AI performance.
https://arxiv.org/abs/1910.05789
MIT License
683 stars 144 forks source link

Train 2 agents simultaneously #139

Closed didacjaquet closed 5 months ago

didacjaquet commented 5 months ago

Hi, Could you explain me if it's possible to train 2 PPO agents simultaneously? And if it's so how to do it.

Thanks

micahcarroll commented 5 months ago

I would check out the PPO tests here. Training with self-play is essentially training 2 PPO agents simultaneously.