Robust Subspace Recovery Layer for Unsupervised Anomaly Detection
This addresses the problem of detecting anomalies in data without labeled examples, which is incremental as it builds on autoencoder-based methods.
The paper tackles unsupervised anomaly detection by proposing a neural network with a robust subspace recovery layer that extracts the underlying subspace from latent representations and removes outliers, achieving state-of-the-art precision and recall in experiments on image and document datasets.
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.