Restricted Generative Projection for One-Class Classification and Anomaly Detection

TL;DR

This paper introduces a simple framework for one-class classification and anomaly detection by learning a mapping to transform normal data into a simple, compact, and informative target distribution, improving anomaly detection accuracy.

Contribution

The paper proposes a novel approach that uses specific simple distributions as targets for transforming normal data, enhancing anomaly detection and one-class classification performance.

Findings

Effective on multiple benchmark datasets

Outperforms baseline methods

Maintains low reconstruction error

Abstract

We present a simple framework for one-class classification and anomaly detection. The core idea is to learn a mapping to transform the unknown distribution of training (normal) data to a known target distribution. Crucially, the target distribution should be sufficiently simple, compact, and informative. The simplicity is to ensure that we can sample from the distribution easily, the compactness is to ensure that the decision boundary between normal data and abnormal data is clear and reliable, and the informativeness is to ensure that the transformed data preserve the important information of the original data. Therefore, we propose to use truncated Gaussian, uniform in hypersphere, uniform on hypersphere, or uniform between hyperspheres, as the target distribution. We then minimize the distance between the transformed data distribution and the target distribution while keeping the reconstruction error for the original data small enough. Comparative studies on multiple benchmark datasets verify the effectiveness of our methods in comparison to baselines.