Are multi-agent environments supported?

IntelLabs / coach

Reinforcement Learning Coach by Intel AI Lab enables easy experimentation with state of the art Reinforcement Learning algorithms

https://intellabs.github.io/coach/

Apache License 2.0

2.32k stars 459 forks source link

Are multi-agent environments supported? #155

Open bbalaji-ucsd opened 5 years ago

bbalaji-ucsd commented 5 years ago

This comment in code indicates that they are supported: https://github.com/NervanaSystems/coach/blob/master/rl_coach/level_manager.py#L50

But don't see any documentation or examples.

gal-leibovich commented 5 years ago

Hi @bbalaji-ucsd,

You are correct, there is infrastructure in place to enable future support in multi-agent RL, but currently it is not officially supported. This is why there are no examples nor documentation. Adding support will probably require a new multi-agent based GraphManager, and potential changes to LevelManager.