Graph similarity learning for change-point detection in dynamic networks

TL;DR

This paper introduces a novel online change-point detection method for dynamic networks using a siamese graph neural network to learn graph similarity, enabling effective detection without prior knowledge of network models.

Contribution

The main contribution is the development of a data-driven, domain-adaptive graph similarity learning approach for real-time change-point detection in diverse dynamic networks.

Findings

Effective detection of change-points in synthetic and real data.

Requires less data history than existing methods.

Applicable to networks with edge weights and node attributes.

Abstract

Dynamic networks are ubiquitous for modelling sequential graph-structured data, e.g., brain connectome, population flows and messages exchanges. In this work, we consider dynamic networks that are temporal sequences of graph snapshots, and aim at detecting abrupt changes in their structure. This task is often termed network change-point detection and has numerous applications, such as fraud detection or physical motion monitoring. Leveraging a graph neural network model, we design a method to perform online network change-point detection that can adapt to the specific network domain and localise changes with no delay. The main novelty of our method is to use a siamese graph neural network architecture for learning a data-driven graph similarity function, which allows to effectively compare the current graph and its recent history. Importantly, our method does not require prior knowledge on the network generative distribution and is agnostic to the type of change-points; moreover, it can be applied to a large variety of networks, that include for instance edge weights and node attributes. We show on synthetic and real data that our method enjoys a number of benefits: it is able to learn an adequate graph similarity function for performing online network change-point detection in diverse types of change-point settings, and requires a shorter data history to detect changes than most existing state-of-the-art baselines.