Using Image Transformations to Learn Network Structure

TL;DR

This paper introduces a method that transforms network data into images, extracts geographic signatures via compression, and uses Bayesian reinforcement learning to improve network planning and decision-making.

Contribution

It presents a novel approach of using image transformations and compression to learn and utilize network structure for reinforcement learning tasks.

Findings

Geographic signatures effectively summarize network structure.

Bayesian reinforcement learning improves decision-making in network planning.

Direct compression-based reinforcement learning is feasible for simple tasks.

Abstract

Many learning tasks require observing a sequence of images and making a decision. In a transportation problem of designing and planning for shipping boxes between nodes, we show how to treat the network of nodes and the flows between them as images. These images have useful structural information that can be statistically summarized. Using image compression techniques, we reduce an image down to a set of numbers that contain interpretable geographic information that we call geographic signatures. Using geographic signatures, we learn network structure that can be utilized to recommend future network connectivity. We develop a Bayesian reinforcement algorithm that takes advantage of statistically summarized network information as priors and user-decisions to reinforce an agent's probabilistic decision. Additionally, we show how reinforcement learning can be used with compression directly without interpretation in simple tasks.