Unifying Local and Global Change Detection in Dynamic Networks

TL;DR

This paper introduces a unified model for detecting both local and global change points in dynamic networks, leveraging a novel generative framework based on MMSB with scalable inference methods.

Contribution

It presents the first unified approach to model and detect local and global changes simultaneously in dynamic networks using a new generative model and scalable inference.

Findings

The model effectively detects change points in synthetic data.

It outperforms baseline methods on real-world network data.

The approach scales to large networks efficiently.

Abstract

Many real-world networks are complex dynamical systems, where both local (e.g., changing node attributes) and global (e.g., changing network topology) processes unfold over time. Local dynamics may provoke global changes in the network, and the ability to detect such effects could have profound implications for a number of real-world problems. Most existing techniques focus individually on either local or global aspects of the problem or treat the two in isolation from each other. In this paper we propose a novel network model that simultaneously accounts for both local and global dynamics. To the best of our knowledge, this is the first attempt at modeling and detecting local and global change points on dynamic networks via a unified generative framework. Our model is built upon the popular mixed membership stochastic blockmodels (MMSB) with sparse co-evolving patterns. We derive an efficient stochastic gradient Langevin dynamics (SGLD) sampler for our proposed model, which allows it to scale to potentially very large networks. Finally, we validate our model on both synthetic and real-world data and demonstrate its superiority over several baselines.