arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2302.05055cs.MAcs.LG
收藏

多智能体强化学习中的低熵通信

Low Entropy Communication in Multi-Agent Reinforcement Learning

Tsinghua University(清华大学)

Lebin Yu, Yunbo Qiu, Qiexiang Wang, Xudong Zhang, Jian Wang

摘要:

多智能体强化学习中的通信因其在协作中的重要作用而受到关注。然而,多智能体系统可能会受到通信资源的限制,因此需要有效的通信技术在现实世界中的场景。根据香农-哈特利定理,在较差信道中可靠传输的消息需要较低的熵。因此,我们的目标是减少多智能体通信中的消息熵。一个基本的挑战是熵的梯度要么是0,要么是无穷大,这使得基于梯度的方法无法使用。为了解决这个问题,我们提出了一个伪梯度下降方案,它通过明智地调整消息的分布来降低熵。我们进行了实验,两个基地的通信框架与六个环境设置,发现我们的计划可以减少消息熵高达90%,几乎没有损失的合作性能。

英文摘要:

Communication in multi-agent reinforcement learning has been drawing attention recently for its significant role in cooperation. However, multi-agent systems may suffer from limitations on communication resources and thus need efficient communication techniques in real-world scenarios. According to the Shannon-Hartley theorem, messages to be transmitted reliably in worse channels require lower entropy. Therefore, we aim to reduce message entropy in multi-agent communication. A fundamental challenge is that the gradients of entropy are either 0 or infinity, disabling gradient-based methods. To handle it, we propose a pseudo gradient descent scheme, which reduces entropy by adjusting the distributions of messages wisely. We conduct experiments on two base communication frameworks with six environment settings and find that our scheme can reduce message entropy by up to 90% with nearly no loss of cooperation performance.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑