Robust Subspace Recovery Layer for Unsupervised Anomaly Detection

ICLR 2020  ·  Chieh-Hsin Lai, Dongmian Zou, Gilad Lerman ·

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. read more

PDF Abstract ICLR 2020 PDF ICLR 2020 Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Unsupervised Anomaly Detection 20NEWS RSRAE AUC (outlier ratio = 0.5) 0.831 # 1
Unsupervised Anomaly Detection Caltech-101 RSRAE AUC (outlier ratio = 0.5) 0.772 # 1
Unsupervised Anomaly Detection Fashion-MNIST RSRAE AUC (outlier ratio = 0.5) 0.833 # 1
Unsupervised Anomaly Detection Reuters-21578 RSRAE AUC (outlier ratio = 0.5) 0.849 # 1

Methods


No methods listed for this paper. Add relevant methods here