search

liuzuxin / DSRL

🔥 Datasets and env wrappers for offline safe reinforcement learning
https://offline-saferl.org
Apache License 2.0
65 stars 4 forks source link

safe multi-agent #2

Closed QingyuanWuNothing closed 3 months ago

QingyuanWuNothing commented 7 months ago

It is excellent work! Whether DSRL support multi-agent settings (e.g., Safe Multi-Agent in safety-gymnasium) in the future?

weakenleg commented 5 months ago

I have the same question, and I hope future work will consider it.

liuzuxin commented 5 months ago

Hi @QingyuanWuNothing @weakenleg , thanks for your interest in our work! My collaborators and I are discussing the best ways of integrating MARL settings with DSRL, but unfortunately, they won't come out in a short timeframe because of the many challenges we might face. BTW you are always welcome to contribute your datasets and environments to this repo.