GLADformer: A Mixed Perspective for Graph-level Anomaly Detection

TL;DR

GLADformer introduces a hybrid graph-level anomaly detection model combining global spectrum-enhanced transformer and spectral GNN modules, effectively capturing both global and local graph anomalies across diverse datasets.

Contribution

The paper presents a novel multi-perspective hybrid model, GLADformer, integrating spectral and spatial features for improved graph-level anomaly detection.

Findings

Outperforms state-of-the-art models on ten real-world datasets.

Effectively captures global and spectral features for anomaly detection.

Demonstrates robustness and generalization across multiple domains.

Abstract

Graph-Level Anomaly Detection (GLAD) aims to distinguish anomalous graphs within a graph dataset. However, current methods are constrained by their receptive fields, struggling to learn global features within the graphs. Moreover, most contemporary methods are based on spatial domain and lack exploration of spectral characteristics. In this paper, we propose a multi-perspective hybrid graph-level anomaly detector namely GLADformer, consisting of two key modules. Specifically, we first design a Graph Transformer module with global spectrum enhancement, which ensures balanced and resilient parameter distributions by fusing global features and spectral distribution characteristics. Furthermore, to uncover local anomalous attributes, we customize a band-pass spectral GNN message passing module that further enhances the model's generalization capability. Through comprehensive experiments on ten real-world datasets from multiple domains, we validate the effectiveness and robustness of GLADformer. This demonstrates that GLADformer outperforms current state-of-the-art models in graph-level anomaly detection, particularly in effectively capturing global anomaly representations and spectral characteristics.