DASVDD: Deep Autoencoding Support Vector Data Descriptor for Anomaly Detection

doi:10.1109/TKDE.2023.3328882

Return

DASVDD: Deep Autoencoding Support Vector Data Descriptor for Anomaly Detection

2024-08-01

Hadi Hojjati *

Narges Armanfard

DOI：10.1109/TKDE.2023.3328882

Original

Original request for help

Save

Abstract

En 中文

One-Class anomaly detection aims to detect anomalies from normal samples using a model trained on normal data. With recent advancements in deep learning, researchers have designed efficient one-class anomaly detection methods. Existing works commonly use neural networks to map the data into a more informative representation and then apply an anomaly detection algorithm. In this paper, we propose a method, DASVDD, that jointly learns the parameters of an autoencoder while minimizing the volume of an enclosing hypersphere on its latent representation. We propose a novel anomaly score that combines the autoencoder's reconstruction error and the distance from the center of the enclosing hypersphere in the latent representation. Minimizing this anomaly score aids us in learning the underlying distribution of the normal class during training. Including the reconstruction error in the anomaly score ensures that DASVDD does not suffer from the hypersphere collapse issue since the DASVDD model does not converge to the trivial solution of mapping all inputs to a constant point in the latent representation. Experimental evaluations on several benchmark datasets show that the proposed method outperforms the commonly used state-of-the-art anomaly detection algorithms while maintaining robust performance across different anomaly classes.

Keywords:

Anomaly detection

deep autoencoder

deep learning

support vector data descriptor

deep autoencoder

deep learning

support vector data descriptor