Continual Model-based Reinforcement Learning for Data Efficient Wireless Network Optimisation

TL;DR

This paper introduces a continual reinforcement learning approach for wireless network optimization, significantly reducing deployment time while maintaining optimization performance.

Contribution

It proposes a novel continual RL method that leverages domain expert knowledge to efficiently adapt control policies across multiple network sites.

Findings

Deployment time halved compared to baseline

Maintains optimization gain without performance drop

Effective in real-world wireless network scenarios

Abstract

We present a method that addresses the pain point of long lead-time required to deploy cell-level parameter optimisation policies to new wireless network sites. Given a sequence of action spaces represented by overlapping subsets of cell-level configuration parameters provided by domain experts, we formulate throughput optimisation as Continual Reinforcement Learning of control policies. Simulation results suggest that the proposed system is able to shorten the end-to-end deployment lead-time by two-fold compared to a reinitialise-and-retrain baseline without any drop in optimisation gain.