Deep Anomaly Detection with Deviation Networks

TL;DR

This paper introduces a neural deviation learning framework for anomaly detection that efficiently leverages limited labeled anomalies to produce superior anomaly scores compared to existing methods.

Contribution

It proposes an end-to-end anomaly scoring method that uses few labeled anomalies and prior probability, improving data efficiency and detection performance.

Findings

Outperforms state-of-the-art methods in anomaly scoring

Requires fewer labeled anomalies for effective detection

Achieves statistically significant deviation of anomalies from normal data

Abstract

Although deep learning has been applied to successfully address many data mining problems, relatively limited work has been done on deep learning for anomaly detection. Existing deep anomaly detection methods, which focus on learning new feature representations to enable downstream anomaly detection methods, perform indirect optimization of anomaly scores, leading to data-inefficient learning and suboptimal anomaly scoring. Also, they are typically designed as unsupervised learning due to the lack of large-scale labeled anomaly data. As a result, they are difficult to leverage prior knowledge (e.g., a few labeled anomalies) when such information is available as in many real-world anomaly detection applications. This paper introduces a novel anomaly detection framework and its instantiation to address these problems. Instead of representation learning, our method fulfills an end-to-end learning of anomaly scores by a neural deviation learning, in which we leverage a few (e.g., multiple to dozens) labeled anomalies and a prior probability to enforce statistically significant deviations of the anomaly scores of anomalies from that of normal data objects in the upper tail. Extensive results show that our method can be trained substantially more data-efficiently and achieves significantly better anomaly scoring than state-of-the-art competing methods.