# GAN Augmented Text Anomaly Detection with Sequences of Deep Statistics

## TL;DR

This paper introduces a two-level framework combining GAN-based data augmentation and deep statistical analysis for effective anomaly detection in sequential discrete data, emphasizing timely identification of rare or unknown anomalies.

## Contribution

It proposes a novel semi-supervised adversarial approach that leverages discriminator layer statistics for anomaly detection in sequences, enhancing detection accuracy for rare anomalies.

## Key findings

- Effective discrimination between in-distribution and out-of-distribution samples.
- Improved anomaly detection performance using deep statistical features.
- Demonstrated applicability to real-world sequential data scenarios.

## Abstract

Anomaly detection is the process of finding data points that deviate from a baseline. In a real-life setting, anomalies are usually unknown or extremely rare. Moreover, the detection must be accomplished in a timely manner or the risk of corrupting the system might grow exponentially. In this work, we propose a two level framework for detecting anomalies in sequences of discrete elements. First, we assess whether we can obtain enough information from the statistics collected from the discriminator's layers to discriminate between out of distribution and in distribution samples. We then build an unsupervised anomaly detection module based on these statistics. As to augment the data and keep track of classes of known data, we lean toward a semi-supervised adversarial learning applied to discrete elements.