Toward Collaborative Reinforcement Learning Agents that Communicate Through Text-Based Natural Language

TL;DR

This paper explores reinforcement learning agents that communicate via natural language text to collaboratively navigate mazes, demonstrating effective communication with limited vocabulary and significant performance improvements.

Contribution

It introduces a novel environment for training RL agents to communicate through natural language, advancing autonomous language-based collaboration.

Findings

Agents achieved a BLEU score of 0.85, indicating effective language use.

Maze completion rate was 100%, outperforming random baselines.

Communication was effective despite imperfect English, enabling successful navigation.

Abstract

Communication between agents in collaborative multi-agent settings is in general implicit or a direct data stream. This paper considers text-based natural language as a novel form of communication between multiple agents trained with reinforcement learning. This could be considered first steps toward a truly autonomous communication without the need to define a limited set of instructions, and natural collaboration between humans and robots. Inspired by the game of Blind Leads, we propose an environment where one agent uses natural language instructions to guide another through a maze. We test the ability of reinforcement learning agents to effectively communicate through discrete word-level symbols and show that the agents are able to sufficiently communicate through natural language with a limited vocabulary. Although the communication is not always perfect English, the agents are still able to navigate the maze. We achieve a BLEU score of 0.85, which is an improvement of 0.61 over randomly generated sequences while maintaining a 100% maze completion rate. This is a 3.5 times the performance of the random baseline using our reference set.