Adaptivity in Agent-Based Routing for Data Networks

TL;DR

This paper explores adaptive multi-agent routing in data networks, showing that RL-suitable action spaces and adaptive interaction structures significantly improve network throughput compared to traditional algorithms.

Contribution

It demonstrates that designing agent action spaces for reinforcement learning and making interaction structures adaptive greatly enhance routing performance.

Findings

RL-compatible action spaces improve routing throughput by over 3.5 times.

Adaptive interaction structures enable agents to realize much of the potential performance gains.

Learning-based routing outperforms standard Bellman-Ford algorithm in simulated network environments.

Abstract

Adaptivity, both of the individual agents and of the interaction structure among the agents, seems indispensable for scaling up multi-agent systems (MAS's) in noisy environments. One important consideration in designing adaptive agents is choosing their action spaces to be as amenable as possible to machine learning techniques, especially to reinforcement learning (RL) techniques. One important way to have the interaction structure connecting agents itself be adaptive is to have the intentions and/or actions of the agents be in the input spaces of the other agents, much as in Stackelberg games. We consider both kinds of adaptivity in the design of a MAS to control network packet routing. We demonstrate on the OPNET event-driven network simulator the perhaps surprising fact that simply changing the action space of the agents to be better suited to RL can result in very large improvements in their potential performance: at their best settings, our learning-amenable router agents achieve throughputs up to three and one half times better than that of the standard Bellman-Ford routing algorithm, even when the Bellman-Ford protocol traffic is maintained. We then demonstrate that much of that potential improvement can be realized by having the agents learn their settings when the agent interaction structure is itself adaptive.