Guarding Graph Neural Networks for Unsupervised Graph Anomaly Detection

TL;DR

This paper introduces G3AD, a framework that enhances unsupervised graph anomaly detection by safeguarding GNNs against anomalies through auxiliary networks and adaptive caching, leading to improved detection performance.

Contribution

The paper proposes a novel G3AD framework that protects GNNs from anomalies in unsupervised settings using correlation constraints and caching, addressing a gap in existing methods.

Findings

G3AD outperforms 20 state-of-the-art methods on synthetic and real datasets.

The framework shows flexible generalization across different GNN architectures.

Extensive experiments validate the effectiveness of the proposed approach.

Abstract

Unsupervised graph anomaly detection aims at identifying rare patterns that deviate from the majority in a graph without the aid of labels, which is important for a variety of real-world applications. Recent advances have utilized Graph Neural Networks (GNNs) to learn effective node representations by aggregating information from neighborhoods. This is motivated by the hypothesis that nodes in the graph tend to exhibit consistent behaviors with their neighborhoods. However, such consistency can be disrupted by graph anomalies in multiple ways. Most existing methods directly employ GNNs to learn representations, disregarding the negative impact of graph anomalies on GNNs, resulting in sub-optimal node representations and anomaly detection performance. While a few recent approaches have redesigned GNNs for graph anomaly detection under semi-supervised label guidance, how to address the adverse effects of graph anomalies on GNNs in unsupervised scenarios and learn effective representations for anomaly detection are still under-explored. To bridge this gap, in this paper, we propose a simple yet effective framework for Guarding Graph Neural Networks for Unsupervised Graph Anomaly Detection (G3AD). Specifically, G3AD first introduces two auxiliary networks along with correlation constraints to guard the GNNs against inconsistent information encoding. Furthermore, G3AD introduces an adaptive caching module to guard the GNNs from directly reconstructing the observed graph data that contains anomalies. Extensive experiments demonstrate that our G3AD can outperform twenty state-of-the-art methods on both synthetic and real-world graph anomaly datasets, with flexible generalization ability in different GNN backbones.