Scalability of Message Encoding Techniques for Continuous Communication Learned with Multi-Agent Reinforcement Learning

TL;DR

This paper examines how message encoding techniques in multi-agent reinforcement learning scale with increasing numbers of agents and information content, comparing mean and attention encoders in a matrix environment.

Contribution

It introduces a comparative analysis of message encoding methods and reveals the superiority of the mean encoder in scalability, along with insights into the communication protocol used.

Findings

Mean message encoder outperforms attention encoder as agent number increases.

Agents use exponential and logarithmic functions to optimize communication.

Communication protocol adapts to preserve important information.

Abstract

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.