Coincident Learning for Unsupervised Anomaly Detection

TL;DR

CoAD introduces a novel unsupervised anomaly detection method leveraging coincident behavior across multi-modal feature slices, improving detection in complex systems with limited labeled data.

Contribution

The paper proposes CoAD, a new unsupervised anomaly detection approach that uses a coincident behavior metric across feature slices, tailored for multi-modal data.

Findings

Outperforms existing methods on real-world datasets

Effective in detecting anomalies with limited labeled data

Validated on synthetic, image, and industrial datasets

Abstract

Anomaly detection is an important task for complex systems (e.g., industrial facilities, manufacturing, large-scale science experiments), where failures in a sub-system can lead to low yield, faulty products, or even damage to components. While complex systems often have a wealth of data, labeled anomalies are typically rare (or even nonexistent) and expensive to acquire. Unsupervised approaches are therefore common and typically search for anomalies either by distance or density of examples in the input feature space (or some associated low-dimensional representation). This paper presents a novel approach called CoAD, which is specifically designed for multi-modal tasks and identifies anomalies based on \textit{coincident} behavior across two different slices of the feature space. We define an \textit{unsupervised} metric, $\hat{F}_\beta$, out of analogy to the supervised classification $F_\beta$ statistic. CoAD uses $\hat{F}_\beta$ to train an anomaly detection algorithm on \textit{unlabeled data}, based on the expectation that anomalous behavior in one feature slice is coincident with anomalous behavior in the other. The method is illustrated using a synthetic outlier data set and a MNIST-based image data set, and is compared to prior state-of-the-art on two real-world tasks: a metal milling data set and a data set from a particle accelerator.