Learning to Ground Decentralized Multi-Agent Communication with Contrastive Learning
This addresses the problem of inefficient communication in multi-agent learning for applications like robotics or gaming, though it is incremental as it builds on existing contrastive learning methods.
The paper tackles the challenge of inducing a common language in decentralized multi-agent systems by treating messages as incomplete views of the environment and maximizing mutual information via contrastive learning, resulting in improved learning performance, speed, and language consistency without extra parameters.
For communication to happen successfully, a common language is required between agents to understand information communicated by one another. Inducing the emergence of a common language has been a difficult challenge to multi-agent learning systems. In this work, we introduce an alternative perspective to the communicative messages sent between agents, considering them as different incomplete views of the environment state. Based on this perspective, we propose a simple approach to induce the emergence of a common language by maximizing the mutual information between messages of a given trajectory in a self-supervised manner. By evaluating our method in communication-essential environments, we empirically show how our method leads to better learning performance and speed, and learns a more consistent common language than existing methods, without introducing additional learning parameters.