QTRAN: Learning to Factorize with Transformation for Cooperative Multi-Agent Reinforcement Learning

TL;DR

QTRAN introduces a flexible factorization method for multi-agent reinforcement learning that overcomes structural constraints of previous models, enabling better performance across diverse cooperative tasks.

Contribution

It proposes a novel, unconstrained factorization approach for joint action-value functions in MARL, expanding the range of solvable cooperative tasks.

Findings

QTRAN outperforms VDN and QMIX in complex tasks.

QTRAN effectively handles non-monotonic and non-additive value functions.

Experimental results show larger margins in penalized cooperative scenarios.

Abstract

We explore value-based solutions for multi-agent reinforcement learning (MARL) tasks in the centralized training with decentralized execution (CTDE) regime popularized recently. However, VDN and QMIX are representative examples that use the idea of factorization of the joint action-value function into individual ones for decentralized execution. VDN and QMIX address only a fraction of factorizable MARL tasks due to their structural constraint in factorization such as additivity and monotonicity. In this paper, we propose a new factorization method for MARL, QTRAN, which is free from such structural constraints and takes on a new approach to transforming the original joint action-value function into an easily factorizable one, with the same optimal actions. QTRAN guarantees more general factorization than VDN or QMIX, thus covering a much wider class of MARL tasks than does previous methods. Our experiments for the tasks of multi-domain Gaussian-squeeze and modified predator-prey demonstrate QTRAN's superior performance with especially larger margins in games whose payoffs penalize non-cooperative behavior more aggressively.