TracInAD: Measuring Influence for Anomaly Detection

TL;DR

This paper introduces TracInAD, a novel influence-based method for anomaly detection on tabular data, leveraging influence measures to improve detection accuracy in unsupervised deep learning models.

Contribution

The paper proposes a new influence-based approach, TracInAD, to enhance anomaly detection in tabular datasets using deep models like Variational Autoencoders.

Findings

Achieves comparable or better detection accuracy than state-of-the-art methods.

Effective influence proxy correlates with anomaly likelihood.

Applicable to medical and cybersecurity datasets.

Abstract

As with many other tasks, neural networks prove very effective for anomaly detection purposes. However, very few deep-learning models are suited for detecting anomalies on tabular datasets. This paper proposes a novel methodology to flag anomalies based on TracIn, an influence measure initially introduced for explicability purposes. The proposed methods can serve to augment any unsupervised deep anomaly detection method. We test our approach using Variational Autoencoders and show that the average influence of a subsample of training points on a test point can serve as a proxy for abnormality. Our model proves to be competitive in comparison with state-of-the-art approaches: it achieves comparable or better performance in terms of detection accuracy on medical and cyber-security tabular benchmark data.