Scalability of Message Encoding Techniques for Continuous Communication Learned with Multi-Agent Reinforcement Learning
This work addresses scalability challenges in multi-agent communication for reinforcement learning systems, though it appears incremental as it compares existing encoding methods.
The paper investigates how increasing message information and agent count affects two encoding methods (mean and attention) in multi-agent reinforcement learning, finding that the mean encoder consistently outperforms the attention encoder on a matrix environment.
Many multi-agent systems require inter-agent communication to properly achieve their goal. By learning the communication protocol alongside the action protocol using multi-agent reinforcement learning techniques, the agents gain the flexibility to determine which information should be shared. However, when the number of agents increases we need to create an encoding of the information contained in these messages. In this paper, we investigate the effect of increasing the amount of information that should be contained in a message and increasing the number of agents. We evaluate these effects on two different message encoding methods, the mean message encoder and the attention message encoder. We perform our experiments on a matrix environment. Surprisingly, our results show that the mean message encoder consistently outperforms the attention message encoder. Therefore, we analyse the communication protocol used by the agents that use the mean message encoder and can conclude that the agents use a combination of an exponential and a logarithmic function in their communication policy to avoid the loss of important information after applying the mean message encoder.