Reconciling Attribute and Structural Anomalies for Improved Graph Anomaly Detection

Abstract

Graph anomaly detection is critical in domains such as healthcare and economics, where identifying deviations can prevent substantial losses. Existing unsupervised approaches strive to learn a single model capable of detecting both attribute and structural anomalies. However, they confront the tug-of-war problem between two distinct types of anomalies, resulting in suboptimal performance. This work presents TripleAD, a mutual distillation-based triple-channel graph anomaly detection framework. It includes three estimation modules to identify the attribute, structural, and mixed anomalies while mitigating the interference between different types of anomalies. In the first channel, we design a multiscale attribute estimation module to capture extensive node interactions and ameliorate the over-smoothing issue. To better identify structural anomalies, we introduce a link-enhanced structure estimation module in the second channel that facilitates information flow to topologically isolated nodes. The third channel is powered by an attribute-mixed curvature, a new indicator that encapsulates both attribute and structural information for discriminating mixed anomalies. Moreover, a mutual distillation strategy is introduced to encourage communication and collaboration between the three channels. Extensive experiments demonstrate the effectiveness of the proposed TripleAD model against strong baselines.

Accepted by IEEE Transactions on Neural Networks and Learning Systems (TNNLS); DOI: https://doi.org/10.1109/TNNLS.2025.3561172

References in corpus (16)

• A Comprehensive Survey on Graph Neural Networks
• Knowledge Distillation: A Survey
• AM-GCN: Adaptive Multi-channel Graph Convolutional Networks
• Anomaly Detection on Attributed Networks via Contrastive Self-Supervised Learning
• Graph Learning: A Survey
• FANG: Leveraging Social Context for Fake News Detection Using Graph Representation
• Missing Data Imputation with Adversarially-trained Graph Convolutional Networks
• Learning on Attribute-Missing Graphs
• Graph Convolutional Networks for Graphs Containing Missing Features
• Self-Supervised Temporal Graph learning with Temporal and Structural Intensity Alignment
• GAD-NR: Graph Anomaly Detection via Neighborhood Reconstruction
• Reconstruction Enhanced Multi-View Contrastive Learning for Anomaly Detection on Attributed Networks
• Interleaved Sequence RNNs for Fraud Detection
• Alleviating Structural Distribution Shift in Graph Anomaly Detection
• Learning Strong Graph Neural Networks with Weak Information
• Motif-Consistent Counterfactuals with Adversarial Refinement for Graph-Level Anomaly Detection