brithbutter

To whom it may concern, In the paper about MAT, it is mentioned that this MARL model is processed within a central executed sequential approach, which is also complex in...

State (cent obs) not used in evaluate actions

1

To whom it may concern, I noticed that in forward() of ma_transformer.py, the state (cent obs) is converted into zeros. However, it seems to be an essential element for critic...

关于多智能体强化学习算法mat的讨论

2

作者您好， 最近阅读了您对于multi-agent transformer 的文章解读，感觉受益匪浅。只是对于mat这个模型有一事不明，还望不吝赐教。根据模型描述看，该模型虽然是多智能体模型，但是顺序地通过transfomer产生每个智能体的决策。这就不免让人疑惑于，这类集中执行（centralized execution）的方法相对于单智能体的强化学习算法是否在时间复杂度上具有优势？这个问题困扰了我很久，还希望得到作者您对此的看法。