Robust Subspace Recovery Layer for Unsupervised Anomaly Detection (2019-03-30T00:00:00.000000Z)

TL;DR

A neural network for unsupervised anomaly detection with a novel robust subspace recovery layer (RSR layer) that seeks to extract the underlying subspace from a latent representation of the given data and removes outliers that lie away from this subspace.

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.

Authors

Chieh-Hsin Lai

1 papers

Dongmian Zou

1 papers

Gilad Lerman

1 papers

TL;DR

Abstract

Authors

References61 items

Generative Adversarial Networks

Robust Subspace Recovery with Adversarial Outliers

q-Space Novelty Detection with Variational Autoencoders

Deep Anomaly Detection Using Geometric Transformations

An Overview of Robust Subspace Recovery

Static and Dynamic Robust PCA and Matrix Completion: A Review

Deep Autoencoding Gaussian Mixture Model for Unsupervised Anomaly Detection

Novelty Detection with GAN

Efficient GAN-Based Anomaly Detection

Clustering and Unsupervised Anomaly Detection with l2 Normalized Deep Auto-Encoder Representations

Robust PCA for Anomaly Detection in Cyber Networks

Future Frame Prediction for Anomaly Detection - A New Baseline

OLE: Orthogonal Low-rank Embedding, A Plug and Play Geometric Loss for Deep Learning

Deep Clustering with Convolutional Autoencoders

Deep Subspace Clustering Networks

Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms

Anomaly Detection with Robust Deep Autoencoders

Wasserstein Generative Adversarial Networks

A Well-Tempered Landscape for Non-convex Robust Subspace Recovery

Robust, Deep and Inductive Anomaly Detection

Improved Training of Wasserstein GANs

Wasserstein GAN

Outlier Detection for Text Data

Deep Structured Energy Based Models for Anomaly Detection

A Comparative Evaluation of Unsupervised Anomaly Detection Algorithms for Multivariate Data

Identity Mappings in Deep Residual Networks

Learning Discriminative Reconstructions for Unsupervised Outlier Removal

Adam: A Method for Stochastic Optimization

ImageNet Large Scale Visual Recognition Challenge

Fast, Robust and Non-convex Subspace Recovery

Auto-Encoding Variational Bayes

Enhancing one-class support vector machines for unsupervised anomaly detection

Isolation-Based Anomaly Detection

Robust Computation of Linear Models by Convex Relaxation

A novel M-estimator for robust PCA

Mining of Massive Datasets

Scikit-learn: Machine Learning in Python

Two proposals for robust PCA using semidefinite programming

Robust PCA via Outlier Pursuit

Median K-Flats for hybrid linear modeling with many outliers

Anomaly detection: A survey

Robust Principal Component Analysis: Exact Recovery of Corrupted Low-Rank Matrices

R1-PCA: rotational invariant L1-norm principal component analysis for robust subspace factorization

The relationship between Precision-Recall and ROC curves

Learning Generative Visual Models from Few Training Examples: An Incremental Bayesian Approach Tested on 101 Object Categories

Quantifying curvelike structures of measures by using L2 Jones quantities

A Framework for Robust Subspace Learning

LOF: identifying density-based local outliers

Support Vector Method for Novelty Detection

NewsWeeder: Learning to Filter Netnews

Rectifiable sets and the Traveling Salesman Problem

2018) as a common approach for RSR

Variational Autoencoder based Anomaly Detection using Reconstruction Probability

Deep Learning

MNIST handwritten digit database

A Novel Anomaly Detection Scheme Based on Principal Component Classifier

Some problems in orthogonal distance and non-orthogonal distance regression

Reuters-21578 Text Categorization Test Collection, Distribution 1.0

Analysis of and on uniformly rectifiable sets

N Figure 14: AUC and AP scores for RSRAE and alternative formulations using Tiny Imagenet (images) and 20

Field of Study

Journal Information

Name

Volume

Venue Information

Name

Type

URL

Alternate Names