对比学习+多智能体强化学习 相关工作总结

leverages contrastive learning to maximize the mutual information between the temporal credits and identity representations of different agents

论文代码相比pymarl的QMIX，只有两处改动：（代码风格不错，而且有注释，赞一个）

arange([start,] stop[, step,], dtype=None) 用于生成等差数列

Learning to Ground Decentralized Multi-Agent Communication with Contrastive Learning

arxiv挂出来的文章，应该还不是最终版

consider the communicative messages sent between agents as different incomplete views of the environment state.

下面这篇笔记分类汇总了我在知乎上分享过的有价值的资料，主要是关于多智能体（深度）强化学习的内容。