Anomaly Detection from a Tensor Train Perspective

TL;DR

This paper introduces tensor train-based algorithms for anomaly detection that leverage data compression to distinguish normal from anomalous data across various datasets.

Contribution

It proposes a novel tensor network approach for anomaly detection, applicable to any tensor network representation, emphasizing data compression of normal versus anomalous structures.

Findings

Effective detection of anomalies in digits, faces, and cybersecurity datasets.

Algorithms successfully preserve normal data structure while isolating anomalies.

Versatile approach applicable to multiple tensor network formats.

Abstract

We present a series of algorithms in tensor networks for anomaly detection in datasets, by using data compression in a Tensor Train representation. These algorithms consist of preserving the structure of normal data in compression and deleting the structure of anomalous data. The algorithms can be applied to any tensor network representation. We test the effectiveness of the methods with digits and Olivetti faces datasets and a cybersecurity dataset to determine cyber-attacks.