Directionality Reinforcement Learning to Operate Multi-Agent System without Communication

TL;DR

This paper introduces directionality reinforcement learning (DRL), a decentralized method enabling multi-agent cooperation without communication or observation, improving efficiency and learning yielding behaviors in maze tasks.

Contribution

The paper proposes a novel DRL approach that enhances multi-agent cooperation without communication, demonstrating its effectiveness through maze experiments.

Findings

DRL outperforms previous methods in maze solving time

Agents learn to yield for others through directionality

DRL achieves multi-agent learning with low costs regardless of agent number

Abstract

This paper establishes directionality reinforcement learning (DRL) technique to propose the complete decentralized multi-agent reinforcement learning method which can achieve cooperation based on each agent's learning: no communication and no observation. Concretely, DRL adds the direction "agents have to learn to reach the farthest goal among reachable ones" to learning agents to operate the agents cooperatively. Furthermore, to investigate the effectiveness of the DRL, this paper compare Q-learning agent with DRL with previous learning agent in maze problems. Experimental results derive that (1) DRL performs better than the previous method in terms of the spending time, (2) the direction makes agents learn yielding action for others, and (3) DRL suggests achieving multiagent learning with few costs for any number of agents.