Classification-Based Anomaly Detection for General Data

ICLR 2020  ·  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.

PDF Abstract ICLR 2020 PDF ICLR 2020 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Anomaly Detection Anomaly Detection on Anomaly Detection on Unlabeled ImageNet-30 vs Flowers-102 GOAD ROC-AUC 92.8 # 4
Network ResNet-18 # 1
Anomaly Detection Anomaly Detection on Unlabeled CIFAR-10 vs LSUN (Fix) GOAD ROC-AUC 78.8 # 8
Network ResNet-18 # 1
Anomaly Detection Anomaly Detection on Unlabeled ImageNet-30 vs CUB-200 GOAD ROC-AUC 90.5 # 3
Network ResNet-18 # 1
Anomaly Detection One-class CIFAR-10 GOAD AUROC 88.2 # 20
Anomaly Detection UEA time-series datasets GOAD Avg. ROC-AUC 87.2 # 2
Anomaly Detection Unlabeled CIFAR-10 vs CIFAR-100 GOAD AUROC 89.2 # 8
Network ResNet-18 # 1

Methods


No methods listed for this paper. Add relevant methods here