Normality-Calibrated Autoencoder for Unsupervised Anomaly Detection on Data Contamination
This addresses the problem of detecting anomalies in data with contamination for applications like fraud detection, though it is incremental as it builds on autoencoder-based methods.
The paper tackles unsupervised anomaly detection on contaminated datasets by proposing a Normality-Calibrated Autoencoder (NCAE) that generates normal samples and maximizes reconstruction errors for anomalies, achieving performance comparable to semi-supervised methods.
In this paper, we propose Normality-Calibrated Autoencoder (NCAE), which can boost anomaly detection performance on the contaminated datasets without any prior information or explicit abnormal samples in the training phase. The NCAE adversarially generates high confident normal samples from a latent space having low entropy and leverages them to predict abnormal samples in a training dataset. NCAE is trained to minimise reconstruction errors in uncontaminated samples and maximise reconstruction errors in contaminated samples. The experimental results demonstrate that our method outperforms shallow, hybrid, and deep methods for unsupervised anomaly detection and achieves comparable performance compared with semi-supervised methods using labelled anomaly samples in the training phase. The source code is publicly available on `https://github.com/andreYoo/NCAE_UAD.git'.