Preprint
Aug 2026
RAD: Rule-Augmented Relational Anomaly Detection
RAD is proposed, a rule-augmented relational anomaly detector that combines heterogeneous graph representation learning with refined symbolic rule signals and improves anomaly ranking over flattened tabular detectors and relational baselines under natural class imbalance.
Noah Dahle, Anne M. Tumlin, N. Tran et al.
· 0 citations