What if my action space is changing every step?

sjtu-marl / malib

A parallel framework for population-based multi-agent reinforcement learning.

https://malib.io

MIT License

498 stars 60 forks source link

What if my action space is changing every step? #6

Closed AI-YULU closed 3 years ago

ying-wen commented 3 years ago

Hi,

Currently, we do not support the dynamic action space. One potential solution for your case is setting one big action space, and masking the invalid actions at each step.

Best, Ying