Fully Independent Communication in Multi-Agent Reinforcement Learning

Several recent works have focused on communication approaches in Multi-Agent Reinforcement Learning (MARL). However, the multiple proposed communication methods might still be too complex and not easily transferable to more practical contexts. One of the reasons is due to the use of the famous parameter sharing trick. In this paper, we investigate how independent learners in MARL that do not share parameters can communicate. We demonstrate that this setting might incur into some problems, to which we propose a new learning scheme as a solution. Our results show that, despite the challenges, independent agents can still learn communication strategies following our method. Additionally, we use this method to investigate how communication in MARL is affected by different network capacities, both for sharing and not sharing parameters.