Online Centralized Non-parametric Change-point Detection via Graph-based Likelihood-ratio Estimation

TL;DR

This paper introduces a kernel-based, graph-smoothness assumption-driven method for real-time detection and localization of change-points in network data streams, improving responsiveness and accuracy.

Contribution

It presents a novel likelihood-ratio estimation approach that leverages graph smoothness to detect change-points and localize affected nodes in a network.

Findings

Effective in synthetic experiments

Accurate change-point localization

Real-time detection capability

Abstract

Consider each node of a graph to be generating a data stream that is synchronized and observed at near real-time. At a change-point $\tau$, a change occurs at a subset of nodes $C$, which affects the probability distribution of their associated node streams. In this paper, we propose a novel kernel-based method to both detect $\tau$ and localize $C$, based on the direct estimation of the likelihood-ratio between the post-change and the pre-change distributions of the node streams. Our main working hypothesis is the smoothness of the likelihood-ratio estimates over the graph, i.e connected nodes are expected to have similar likelihood-ratios. The quality of the proposed method is demonstrated on extensive experiments on synthetic scenarios.