Universal Data Anomaly Detection via Inverse Generative Adversary Network
This addresses data anomaly detection for applications where only anomaly-free historical samples are available, but it appears incremental as it builds on existing generative adversarial network concepts.
The paper tackles the problem of detecting data anomalies without training data for anomalous distributions by proposing a semi-supervised deep learning technique based on an inverse generative adversary network, achieving detection under a composite alternative hypothesis where measurements are from an unknown distribution with positive distance from the null hypothesis.
The problem of detecting data anomaly is considered. Under the null hypothesis that models anomaly-free data, measurements are assumed to be from an unknown distribution with some authenticated historical samples. Under the composite alternative hypothesis, measurements are from an unknown distribution positive distance away from the distribution under the null hypothesis. No training data are available for the distribution of anomaly data. A semi-supervised deep learning technique based on an inverse generative adversary network is proposed.