Studies on the GAN-Based Anomaly Detection Methods for the Time Series Data

Anomaly detection (AD) for times series data using the generative adversarial network (GAN) has been proposed in recent years. According to previous study, GAN-based AD outperformed cumulative sum (CUSUM) chart. However, no framework comparison is provided their works. So, we conduct new studies crucial methods (the MAD-GAN and TAnoGAN). First, propose a fair systematic comparisons prediction performance of as well evaluate three with four simulation secure water treatment system data. Under framework, CUSUM chart generally shows performances better than methods. Our results imply that more follow-up are required before deploying Second, find adjusting number backpropagation steps inverse mapping technique can improve Furthermore, monitoring residuals fitted model significantly improves