Structured and Unstructured Outlier Identification for Robust PCA: A Non iterative, Parameter free Algorithm

Abstract

Robust PCA, the problem of PCA in the presence of outliers has been extensively investigated in the last few years. Here we focus on Robust PCA in the outlier model where each column of the data matrix is either an inlier or an outlier. Most of the existing methods for this model assumes either the knowledge of the dimension of the lower dimensional subspace or the fraction of outliers in the system. However in many applications knowledge of these parameters is not available. Motivated by this we propose a parameter free outlier identification method for robust PCA which a) does not require the knowledge of outlier fraction, b) does not require the knowledge of the dimension of the underlying subspace, c) is computationally simple and fast d) can handle structured and unstructured outliers. Further, analytical guarantees are derived for outlier identification and the performance of the algorithm is compared with the existing state of the art methods in both real and synthetic data for various outlier structures.

arXiv commentarXiv admin note: substantial text overlap with arXiv:1804.04791

References in corpus (19)

• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• A survey of cross-validation procedures for model selection
• A Tutorial on Principal Component Analysis
• The Augmented Lagrange Multiplier Method for Exact Recovery of Corrupted Low-Rank Matrices
• Rank-Sparsity Incoherence for Matrix Decomposition
• A geometric analysis of subspace clustering with outliers
• Fast Algorithms for Robust PCA via Gradient Descent
• Coherence Pursuit: Fast, Simple, and Robust Principal Component Analysis
• Distributions of Angles in Random Packing on Spheres
• An Overview of Robust Subspace Recovery
• A Novel M-Estimator for Robust PCA
• Static and Dynamic Robust PCA and Matrix Completion: A Review
• Stable Principal Component Pursuit
• Fast, Robust and Non-convex Subspace Recovery
• Noise Statistics Oblivious GARD For Robust Regression With Sparse Outliers
• Signal and Noise Statistics Oblivious Orthogonal Matching Pursuit
• Thresholding based Efficient Outlier Robust PCA
• Path Thresholding: Asymptotically Tuning-Free High-Dimensional Sparse Regression
• Fast, Parameter free Outlier Identification for Robust PCA

Cited by in corpus (3)

• Subspace clustering without knowing the number of clusters: A parameter free approach
• A robust principal component analysis for outlier identification in messy microcalorimeter data
• Tune smarter not harder: A principled approach to tuning learning rates for shallow nets