Self-Discriminative Modeling for Anomalous Graph Detection

Identifying anomalous graphs is essential in real-world scenarios such as molecular and social network analysis, yet anomalous samples are generally scarce and unavailable. This paper proposes a Self-Discriminative Modeling (SDM) framework that trains a deep neural network only on normal graphs to detect anomalous graphs. The neural network simultaneously learns to construct pseudo-anomalous graphs from normal graphs and learns an anomaly detector to recognize these pseudo-anomalous graphs. As a result, these pseudo-anomalous graphs interpolate between normal graphs and real anomalous graphs, which leads to a reliable decision boundary of anomaly detection. In this framework, we develop three algorithms with different computational efficiencies and stabilities for anomalous graph detection. Extensive experiments on 12 different graph benchmarks demonstrated that the three variants of SDM consistently outperform the state-of-the-art GLAD baselines. The success of our methods stems from the integration of the discriminative classifier and the well-posed pseudo-anomalous graphs, which provided new insights for graph-level anomaly detection.

Graphs are widely used to represent complex relationships, such as connections in social networks or molecular interactions in chemistry. Sometimes, unusual or anomalous graphs appear, which indicate problems like fraudulent activity or unique chemical properties. Detecting these anomalous graphs is crucial, yet it is challenging because they are rare and often unknown in advance. Our research introduces a graph-level anomaly detection method, which aims to detect anomalous graphs by only studying normal graphs. We achieve this via generating auxiliary examples of slight disturbances on the normal graphs, creating learning opportunities to distinguish normal graphs from these "pseudo" anomalies. These self-generated examples help the model refine a more reliable decision boundary between normal and anomalous data. Our experiments show that our method significantly outperforms current state-of-the-art baselines, offering a new insight to detect these rare but critical anomalies in various real-world scenarios.