Post-Hoc Calibrated Anomaly Detection
This work addresses the challenge of accurate anomaly detection in machine learning, offering incremental improvements in calibration techniques for models trained with outlier exposure.
The paper tackles the problem of improving anomaly detection and localization by applying post-hoc calibration methods, such as Platt scaling and Beta calibration, to deep unsupervised models, achieving enhanced results through techniques like gradient-based input perturbation and using synthesized spectral data for calibration.
Deep unsupervised anomaly detection has seen improvements in a supervised binary classification paradigm in which auxiliary external data is included in the training set as anomalous data in a process referred to as outlier exposure, which opens the possibility of exploring the efficacy of post-hoc calibration for anomaly detection and localization. Post-hoc Platt scaling and Beta calibration are found to improve results with gradient-based input perturbation, as well as post-hoc training with a strictly proper loss of a base model initially trained on an unsupervised loss. Post-hoc calibration is also found at times to be more effective using random synthesized spectral data as labeled anomalous data in the calibration set, suggesting that outlier exposure is superior only for initial training.