TAnoGAN: Time Series Anomaly Detection with Generative Adversarial Networks

TL;DR

TAnoGan is a novel unsupervised GAN-based method for detecting anomalies in time series data, especially effective with limited data points, outperforming traditional and neural network models across diverse real-world datasets.

Contribution

The paper introduces TAnoGan, a new GAN-based approach specifically designed for anomaly detection in time series with small datasets, filling a gap in existing methods.

Findings

TAnoGan outperforms traditional anomaly detection methods.

TAnoGan performs better than existing neural network models.

Effective across various real-world time series datasets.

Abstract

Anomaly detection in time series data is a significant problem faced in many application areas such as manufacturing, medical imaging and cyber-security. Recently, Generative Adversarial Networks (GAN) have gained attention for generation and anomaly detection in image domain. In this paper, we propose a novel GAN-based unsupervised method called TAnoGan for detecting anomalies in time series when a small number of data points are available. We evaluate TAnoGan with 46 real-world time series datasets that cover a variety of domains. Extensive experimental results show that TAnoGan performs better than traditional and neural network models.