Towards Language-Augmented Multi-Agent Deep Reinforcement Learning

TL;DR

This paper explores grounding multi-agent reinforcement learning in human language, showing that language-augmented agents achieve better performance, interpretability, and generalization compared to emergent communication methods.

Contribution

It introduces a framework for training agents to produce and interpret natural language, enhancing communication efficiency and interpretability in multi-agent systems.

Findings

Language grounding improves agent performance across tasks.

Language-augmented agents develop more informative internal representations.

Grounded agents generalize better to new partners and facilitate human interaction.

Abstract

Most prior works on communication in multi-agent reinforcement learning have focused on emergent communication, which often results in inefficient and non-interpretable systems. Inspired by the role of language in natural intelligence, we investigate how grounding agents in a human-defined language can improve the learning and coordination of embodied agents. We propose a framework in which agents are trained not only to act but also to produce and interpret natural language descriptions of their observations. This language-augmented learning serves a dual role: enabling efficient and interpretable communication between agents, and guiding representation learning. We demonstrate that language-augmented agents outperform emergent communication baselines across various tasks. Our analysis reveals that language grounding leads to more informative internal representations, better generalization to new partners, and improved capability for human-agent interaction. These findings demonstrate the effectiveness of integrating structured language into multi-agent learning and open avenues for more interpretable and capable multi-agent systems.