Robust Subspace Recovery Layer for Unsupervised Anomaly Detection

Abstract

We propose a neural network for unsupervised anomaly detection with a novel robust subspace recovery layer (RSR layer). This layer seeks to extract the underlying subspace from a latent representation of the given data and removes outliers that lie away from this subspace. It is used within an autoencoder. The encoder maps the data into a latent space, from which the RSR layer extracts the subspace. The decoder then smoothly maps back the underlying subspace to a "manifold" close to the original inliers. Inliers and outliers are distinguished according to the distances between the original and mapped positions (small for inliers and large for outliers). Extensive numerical experiments with both image and document datasets demonstrate state-of-the-art precision and recall.

This work is on the ICLR 2020 conference

References in corpus (7)

• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• Efficient GAN-Based Anomaly Detection
• Deep Anomaly Detection Using Geometric Transformations
• An Overview of Robust Subspace Recovery
• Clustering and Unsupervised Anomaly Detection with L2 Normalized Deep Auto-Encoder Representations
• Robust PCA for Anomaly Detection in Cyber Networks
• Robust Subspace Recovery with Adversarial Outliers

Cited by in corpus (8)

• Generalized Out-of-Distribution Detection: A Survey
• Adaptive Model Pooling for Online Deep Anomaly Detection from a Complex Evolving Data Stream
• Unsupervised Outlier Detection using Memory and Contrastive Learning
• An Iterative Method for Unsupervised Robust Anomaly Detection Under Data Contamination
• The Sound Demixing Challenge 2023 $\unicode{x2013}$ Music Demixing Track
• GRAM: An Interpretable Approach for Graph Anomaly Detection using Gradient Attention Maps
• Towards an Awareness of Time Series Anomaly Detection Models' Adversarial Vulnerability
• Unsupervised Deep One-Class Classification with Adaptive Threshold based on Training Dynamics