Deep Anomaly Detection under Labeling Budget Constraints

TL;DR

This paper introduces a theoretical framework and a new semi-supervised learning approach for anomaly detection that optimizes data labeling under budget constraints, achieving state-of-the-art results across multiple data types.

Contribution

It provides theoretical conditions for anomaly score generalization and proposes an optimal data labeling strategy combined with a semi-supervised learning framework.

Findings

Achieves state-of-the-art semi-supervised anomaly detection performance

Develops a theoretical understanding of anomaly score generalization

Proposes an optimal labeling strategy under budget constraints

Abstract

Selecting informative data points for expert feedback can significantly improve the performance of anomaly detection (AD) in various contexts, such as medical diagnostics or fraud detection. In this paper, we determine a set of theoretical conditions under which anomaly scores generalize from labeled queries to unlabeled data. Motivated by these results, we propose a data labeling strategy with optimal data coverage under labeling budget constraints. In addition, we propose a new learning framework for semi-supervised AD. Extensive experiments on image, tabular, and video data sets show that our approach results in state-of-the-art semi-supervised AD performance under labeling budget constraints.