Improving Generalizability of Graph Anomaly Detection Models via Data Augmentation
arXiv:2306.10534 · doi:10.1109/TKDE.2023.3271771
Abstract
Graph anomaly detection (GAD) is a vital task since even a few anomalies can pose huge threats to benign users. Recent semi-supervised GAD methods, which can effectively leverage the available labels as prior knowledge, have achieved superior performances than unsupervised methods. In practice, people usually need to identify anomalies on new (sub)graphs to secure their business, but they may lack labels to train an effective detection model. One natural idea is to directly adopt a trained GAD model to the new (sub)graph for testing. However, we find that existing semi-supervised GAD methods suffer from poor generalization issue, i.e., well-trained models could not perform well on an unseen area (i.e., not accessible in training) of the same graph. It may cause great troubles. In this paper, we base on the phenomenon and propose a general and novel research problem of generalized graph anomaly detection that aims to effectively identify anomalies on both the training-domain graph and unseen testing graph to eliminate potential dangers. Nevertheless, it is a challenging task since only limited labels are available, and the normal background may differ between training and testing data. Accordingly, we propose a data augmentation method named \textit{AugAN} (\uline{Aug}mentation for \uline{A}nomaly and \uline{N}ormal distributions) to enrich training data and boost the generalizability of GAD models. Experiments verify the effectiveness of our method in improving model generalizability.
arXiv commentAccepted to IEEE Transactions on Knowledge and Data Engineering (TKDE). arXiv admin note: substantial text overlap with arXiv:2209.10168
References in corpus (22)
- Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
- Prototypical Networks for Few-shot Learning
- Deep Learning for Anomaly Detection: A Review
- Generalizing from a Few Examples: A Survey on Few-Shot Learning
- GCC: Graph Contrastive Coding for Graph Neural Network Pre-Training
- Anomaly Detection on Attributed Networks via Contrastive Self-Supervised Learning
- Domain Generalization via Model-Agnostic Learning of Semantic Features
- A Semi-supervised Graph Attentive Network for Financial Fraud Detection
- Generalizing to Unseen Domains via Adversarial Data Augmentation
- GraphSMOTE: Imbalanced Node Classification on Graphs with Graph Neural Networks
- Batch-Instance Normalization for Adaptively Style-Invariant Neural Networks
- One-Class Graph Neural Networks for Anomaly Detection in Attributed Networks
- Generalizing to Unseen Domains: A Survey on Domain Generalization
- Graph-based Anomaly Detection and Description: A Survey
- Rethinking Graph Neural Networks for Anomaly Detection
- Anomaly Detection based on Zero-Shot Outlier Synthesis and Hierarchical Feature Distillation
- BOND: Benchmarking Unsupervised Outlier Node Detection on Static Attributed Graphs
- Multi-scale Anomaly Detection on Attributed Networks
- Handling Distribution Shifts on Graphs: An Invariance Perspective
- Data Augmentation for Deep Graph Learning: A Survey
- Action Sequence Augmentation for Early Graph-based Anomaly Detection
- Graph-Relational Domain Adaptation