Towards Effective Federated Graph Anomaly Detection via Self-boosted Knowledge Distillation

Jinyu Cai, Yunhe Zhang, Zhoumin Lu, Wenzhong Guo, See-Kiong Ng

Abstract: Graph anomaly detection (GAD) aims to identify anomalous graphs that significantly deviate from other ones, which has raised growing attention due to the broad existence and complexity of graph-structured data in many real-world scenarios. However, existing GAD methods usually execute with centralized training, which may lead to privacy leakage risk in some sensitive cases, thereby impeding collaboration among organizations seeking to collectively develop robust GAD models. Although federated learning offers a promising solution, the prevalent non-IID problems and high communication costs present significant challenges, particularly pronounced in collaborations with graph data distributed among different participants. To tackle these challenges, we propose an effective federated graph anomaly detection framework (FGAD). We first introduce an anomaly generator to perturb the normal graphs to be anomalous and train a powerful anomaly detector by distinguishing generated anomalous graphs from normal ones. We subsequently leverage a student model to distill knowledge from the trained anomaly detector (teacher model), which aims to maintain the personality of local models and alleviate the adverse impact of non-IID problems. Additionally, we design an effective collaborative learning mechanism that facilitates the personalization preservation of local models and significantly reduces communication costs among clients. Empirical results of diverse GAD tasks demonstrate the superiority and efficiency of FGAD.

Primary Subject Area: [Content] Multimodal Fusion

Secondary Subject Area: [Experience] Multimedia Applications

Relevance To Conference: This work contributes to multimedia/multimodal processing from the following aspects: (1) The proposed FGAD is able to effectively process multi-dataset anomaly detection from varied sources, e.g., social networks, biological, and chemical domains, thus enhancing its adaptability and utility in diverse multimedia contexts. (2) By facilitating anomaly detection in a federated learning framework, FGAD significantly mitigates privacy concerns and enables secure data analysis in environments where data sharing is restricted, which is crucial for maintaining privacy and security in multimedia processing. (3) FGAD improves scalability and efficiency, which is essential for large-scale multimedia/multimodal applications. In a nutshell, our work broadens the scope of anomaly detection and shows great potential for handling multimedia/multimodal applications in real-world scenarios.

Submission Number: 3845