论文标题
学习通过对比度学习地面分散的多代理沟通
Learning to Ground Decentralized Multi-Agent Communication with Contrastive Learning
论文作者
论文摘要
为了使沟通成功地发生，代理之间需要一种通用语言才能了解彼此传达的信息。诱导通用语言的出现一直是多机构学习系统的艰巨挑战。在这项工作中，我们对代理之间发送的交流消息介绍了另一种观点，将其视为环境状态的不同观点。基于这一观点，我们提出了一种简单的方法来诱导通用语言的出现，通过以自我监督的方式最大化给定轨迹的消息之间的相互信息。通过评估我们在沟通环境中的方法，我们从经验上展示了我们的方法如何带来更好的学习绩效和速度，并在没有引入其他学习参数的情况下学习了比现有方法更一致的通用语言。
For communication to happen successfully, a common language is required between agents to understand information communicated by one another. Inducing the emergence of a common language has been a difficult challenge to multi-agent learning systems. In this work, we introduce an alternative perspective to the communicative messages sent between agents, considering them as different incomplete views of the environment state. Based on this perspective, we propose a simple approach to induce the emergence of a common language by maximizing the mutual information between messages of a given trajectory in a self-supervised manner. By evaluating our method in communication-essential environments, we empirically show how our method leads to better learning performance and speed, and learns a more consistent common language than existing methods, without introducing additional learning parameters.
