CoDreamer: Communication-Based Decentralised World Models
This work addresses sample efficiency for multi-agent RL, but it is incremental as it builds on the existing Dreamer algorithm.
The paper tackles sample efficiency in multi-agent reinforcement learning by extending the Dreamer algorithm to multi-agent environments with a two-level communication system using Graph Neural Networks, demonstrating superiority over baseline methods.
Sample efficiency is a critical challenge in reinforcement learning. Model-based RL has emerged as a solution, but its application has largely been confined to single-agent scenarios. In this work, we introduce CoDreamer, an extension of the Dreamer algorithm for multi-agent environments. CoDreamer leverages Graph Neural Networks for a two-level communication system to tackle challenges such as partial observability and inter-agent cooperation. Communication is separately utilised within the learned world models and within the learned policies of each agent to enhance modelling and task-solving. We show that CoDreamer offers greater expressive power than a naive application of Dreamer, and we demonstrate its superiority over baseline methods across various multi-agent environments.