Liron Bergman, Yedid Hoshen
Anomaly detection, finding patterns that substantially deviate from those seen previously, is one of the fundamental problems of artificial intelligence. Recently, classification-based methods were shown to achieve superior results on this task. In this work, we present a unifying view and propose an open-set method, GOAD, to relax current generalization assumptions. Furthermore, we extend the applicability of transformation-based methods to non-image data using random affine transformations. Our method is shown to obtain state-of-the-art accuracy and is applicable to broad data types. The strong performance of our method is extensively validated on multiple datasets from different domains.
| Task | Dataset | Metric | Value | Model |
|---|---|---|---|---|
| Anomaly Detection | Anomaly Detection on Anomaly Detection on Unlabeled ImageNet-30 vs Flowers-102 | ROC-AUC | 92.8 | GOAD |
| Anomaly Detection | Anomaly Detection on Unlabeled CIFAR-10 vs LSUN (Fix) | ROC-AUC | 78.8 | GOAD |
| Anomaly Detection | UEA time-series datasets | Avg. ROC-AUC | 87.2 | GOAD |
| Anomaly Detection | Unlabeled CIFAR-10 vs CIFAR-100 | AUROC | 89.2 | GOAD |
| Anomaly Detection | Anomaly Detection on Unlabeled ImageNet-30 vs CUB-200 | ROC-AUC | 90.5 | GOAD |
| Anomaly Detection | One-class CIFAR-10 | AUROC | 88.2 | GOAD |