寇谦

RLHF in MARL

2

I want to konw if there are any researches focusing on leveraging RLHF in MARL problems.

训练的simple_tag图不对

3


Hello! The class MediumLevelPlanner used in overcooked.py seems to be deprecated in the latest overcooked_ai version, which class can be used instead?

The confusing transformation about rewards to rtgs.

1

In the get_episode() function, the rewards have been turned into reward-to-gos, which is not describe in the paper. for agent_trajectory in episode: rtgs = 0 for i in reversed(range(len(agent_trajectory))): rtgs...

真实面试题缺失

1

如题

Could you please provide the api server.py for chunk convert