Unsupervised Learning for Anomaly Detection: A Comparison of Deep Generative Models.

Kitgak Simon

Keywords: Detection, Unsupervised, Learning, Anomaly, Deep, Regenerative, Models

TL;DR: Unsupervised Learning for Anomaly Detection

Abstract: Anomaly detection is a critical task in various domains, including cybersecurity, fraud detection, and health monitoring. Traditional methods for anomaly detection rely on handcrafted features and require expert knowledge, which can be time-consuming and expensive. Recently, deep generative models have shown promise for unsupervised anomaly detection. In this paper, we compare the performance of various deep generative models, including variational autoencoders, generative adversarial networks, and flow-based models, for anomaly detection on several benchmark datasets.