A Multi-Agent Multi-Environment Mixed Q-Learning for Partially Decentralized Wireless Network Optimization

TL;DR

This paper introduces a novel multi-agent mixed Q-learning algorithm for partially decentralized wireless networks, improving learning speed and efficiency while maintaining low error rates, addressing limitations of existing centralized methods.

Contribution

It extends multi-environment mixed Q-learning to multi-agent, partially decentralized wireless networks, enabling scalable, efficient learning with limited information sharing.

Findings

50% faster than centralized MEMQ with 20% higher APE

25% faster than other decentralized Q-learning algorithms with 40% less APE

Convergence of the proposed multi-agent MEMQ demonstrated

Abstract

Q-learning is a powerful tool for network control and policy optimization in wireless networks, but it struggles with large state spaces. Recent advancements, like multi-environment mixed Q-learning (MEMQ), improves performance and reduces complexity by integrating multiple Q-learning algorithms across multiple related environments so-called digital cousins. However, MEMQ is designed for centralized single-agent networks and is not suitable for decentralized or multi-agent networks. To address this challenge, we propose a novel multi-agent MEMQ algorithm for partially decentralized wireless networks with multiple mobile transmitters (TXs) and base stations (BSs), where TXs do not have access to each other's states and actions. In uncoordinated states, TXs act independently to minimize their individual costs. In coordinated states, TXs use a Bayesian approach to estimate the joint state based on local observations and share limited information with leader TX to minimize joint cost. The cost of information sharing scales linearly with the number of TXs and is independent of the joint state-action space size. The proposed scheme is 50% faster than centralized MEMQ with only a 20% increase in average policy error (APE) and is 25% faster than several advanced decentralized Q-learning algorithms with 40% less APE. The convergence of the algorithm is also demonstrated.