Dual Ensembled Multiagent Q-Learning with Hypernet Regularizer

TL;DR

This paper introduces a novel multiagent Q-learning algorithm that reduces overestimation by combining ensemble methods with a hypernet regularizer, improving stability and performance in multiagent environments.

Contribution

It proposes a dual ensemble approach with a hypernet regularizer to address overestimation in both target estimation and online optimization in multiagent Q-learning.

Findings

Effectively reduces overestimation in multiagent Q-learning.

Improves stability and performance across multiple multiagent benchmarks.

Demonstrates superior results compared to existing methods.

Abstract

Overestimation in single-agent reinforcement learning has been extensively studied. In contrast, overestimation in the multiagent setting has received comparatively little attention although it increases with the number of agents and leads to severe learning instability. Previous works concentrate on reducing overestimation in the estimation process of target Q-value. They ignore the follow-up optimization process of online Q-network, thus making it hard to fully address the complex multiagent overestimation problem. To solve this challenge, in this study, we first establish an iterative estimation-optimization analysis framework for multiagent value-mixing Q-learning. Our analysis reveals that multiagent overestimation not only comes from the computation of target Q-value but also accumulates in the online Q-network's optimization. Motivated by it, we propose the Dual Ensembled Multiagent Q-Learning with Hypernet Regularizer algorithm to tackle multiagent overestimation from two aspects. First, we extend the random ensemble technique into the estimation of target individual and global Q-values to derive a lower update target. Second, we propose a novel hypernet regularizer on hypernetwork weights and biases to constrain the optimization of online global Q-network to prevent overestimation accumulation. Extensive experiments in MPE and SMAC show that the proposed method successfully addresses overestimation across various tasks.