期刊名称：Journal of Artificial Intelligence and Soft Computing Research
电子版ISSN：2083-2567
出版年度：2020
卷号：10
期号：3
页码：189-207
DOI：10.2478/jaiscr-2020-0013
出版社：Walter de Gruyter GmbH
摘要：We consider the problem of multi agents cooperating in a partially-observable environment. Agents must learn to coordinate and share relevant information to solve the tasks successfully. This article describes Asynchronous Advantage Actor-Critic with Communication (A3C2), an end-to-end differentiable approach where agents learn policies and communication protocols simultaneously. A3C2 uses a centralized learning, distributed execution paradigm, supports independent agents, dynamic team sizes, partially-observable environments, and noisy communications. We compare and show that A3C2 outperforms other state-of-the-art proposals in multiple environments.
关键词：multi-agent systems ; deep reinforcement learning ; centralized learning
