Computing effective communication policies in multiagent systems
Proceedings of the 6th international joint conference on Autonomous agents and multiagent systems
Perseus: randomized point-based value iteration for POMDPs
Journal of Artificial Intelligence Research
Hi-index | 0.00 |
Communication is a key for facilitating multi-agent coordination on cooperative problems. On unknown problems, however, it is hard to construct beneficial communication codes. In order to tackle such problems, we focus on a method that allows agents to learn communication codes autonomously. Kasai et al. [2] proposed Signal Learning, by which agents learn policies of communication and action concurrently in multi-agent reinforcement learning framework. In this paper, we extend the existing signal learning and apply the extended method to an example problem, where agents can observe only partial information, for verifying the power of communication. We show that the performance of the proposed method is better than that of the existing method, and agents can obtain optimal policies on the applied problem by using the proposed method.