Graph Convolutional Reinforcement Learning for Collaborative Queuing Agents

TL;DR

This paper introduces a graph convolutional multi-agent reinforcement learning method to optimize network flow scheduling, significantly improving throughput and delay performance in various scenarios.

Contribution

It presents a novel DGN-based multi-agent reinforcement learning approach for cooperative network flow management, outperforming traditional centralized and distributed methods.

Findings

DGN-based approach effectively meets throughput and delay requirements.

Cooperative agents outperform non-cooperative benchmarks.

Method adapts well across different network scenarios.

Abstract

In this paper, we explore the use of multi-agent deep learning as well as learning to cooperate principles to meet stringent service level agreements, in terms of throughput and end-to-end delay, for a set of classified network flows. We consider agents built on top of a weighted fair queuing algorithm that continuously set weights for three flow groups: gold, silver, and bronze. We rely on a novel graph-convolution based, multi-agent reinforcement learning approach known as DGN. As benchmarks, we propose centralized and distributed deep Q-network approaches and evaluate their performances in different network, traffic, and routing scenarios, highlighting the effectiveness of our proposals and the importance of agent cooperation. We show that our DGN-based approach meets stringent throughput and delay requirements across all scenarios.