Graph Regularized Autoencoder and its Application in Unsupervised Anomaly Detection
This work addresses the challenge of effective dimensionality reduction for high-dimensional data in unsupervised learning tasks like anomaly detection, offering a novel method that enhances performance in specific domains.
The authors tackled the problem of dimensionality reduction for unsupervised anomaly detection by proposing a graph regularized autoencoder that uses a minimum spanning tree (MST) to approximate geodesic distances, which outperformed alternative methods across 20 benchmark datasets and improved performance in generative adversarial networks and clustering applications.
Dimensionality reduction is a crucial first step for many unsupervised learning tasks including anomaly detection and clustering. Autoencoder is a popular mechanism to accomplish dimensionality reduction. In order to make dimensionality reduction effective for high-dimensional data embedding nonlinear low-dimensional manifold, it is understood that some sort of geodesic distance metric should be used to discriminate the data samples. Inspired by the success of geodesic distance approximators such as ISOMAP, we propose to use a minimum spanning tree (MST), a graph-based algorithm, to approximate the local neighborhood structure and generate structure-preserving distances among data points. We use this MST-based distance metric to replace the Euclidean distance metric in the embedding function of autoencoders and develop a new graph regularized autoencoder, which outperforms a wide range of alternative methods over 20 benchmark anomaly detection datasets. We further incorporate the MST regularizer into two generative adversarial networks and find that using the MST regularizer improves the performance of anomaly detection substantially for both generative adversarial networks. We also test our MST regularized autoencoder on two datasets in a clustering application and witness its superior performance as well.