Unsupervised Deep One-Class Classification with Adaptive Threshold based on Training Dynamics
Kim, Minkyung, Kim, Junsik, Yu, Jongmin, Choi, Jun Kyun
–arXiv.org Artificial Intelligence
One-class classification has been a prevailing method in building deep anomaly detection models under the assumption that a dataset consisting of normal samples is available. In practice, however, abnormal samples are often mixed in a training dataset, and they detrimentally affect the training of deep models, which limits their applicability. For robust normality learning of deep practical models, we propose an unsupervised deep one-class classification that learns normality from pseudo-labeled normal samples, i.e., outlier detection in single cluster scenarios. To this end, we propose a pseudo-labeling method by an adaptive threshold selected by ranking-based training dynamics. The experiments on 10 anomaly detection benchmarks show that our method effectively improves performance on anomaly detection by sizable margins.
arXiv.org Artificial Intelligence
Feb-12-2023
- Country:
- North America > United States (0.04)
- Europe > United Kingdom
- England > Greater London > London (0.04)
- Genre:
- Research Report (0.40)
- Technology: