History-based Anomaly Detector: an Adversarial Approach to Anomaly Detection

TL;DR

This paper introduces HistoryAD, a novel adversarial anomaly detection method that leverages training history of GANs to identify normal versus anomalous data, achieving top-tier results across multiple datasets.

Contribution

It proposes a new self-supervised adversarial approach using training history of GANs for improved anomaly detection performance.

Findings

Achieves top-tier results on several datasets.

Outperforms several state-of-the-art methods.

Demonstrates effectiveness of training history in anomaly detection.

Abstract

Anomaly detection is a difficult problem in many areas and has recently been subject to a lot of attention. Classifying unseen data as anomalous is a challenging matter. Latest proposed methods rely on Generative Adversarial Networks (GANs) to estimate the normal data distribution, and produce an anomaly score prediction for any given data. In this article, we propose a simple yet new adversarial method to tackle this problem, denoted as History-based anomaly detector (HistoryAD). It consists of a self-supervised model, trained to recognize 'normal' samples by comparing them to samples based on the training history of a previously trained GAN. Quantitative and qualitative results are presented evaluating its performance. We also present a comparison to several state-of-the-art methods for anomaly detection showing that our proposal achieves top-tier results on several datasets.