Learning Robust and Discriminative Subspace With Low-Rank Constraints
Abstract
In this paper, we aim at learning robust and discriminative subspaces from noisy data. Subspace learning is widely used in extracting discriminative features for classifica- tion. However, when data are contaminated with severe noise, the performance of most existing subspace learning methods would be limited. Recent advances in low-rank modeling provide effective solutions for removing noise or outliers contained in sample sets, which motivates us to take advantage of low-rank constraints in order to exploit robust and discriminative subspace for classification. In particular, we present a discriminative subspace learning method called the supervised regularization- based robust subspace (SRRS) approach, by incorporating the low-rank constraint. SRRS seeks low-rank representations from the noisy data, and learns a discriminative subspace from the recovered clean data jointly. A supervised regularization function is designed to make use of the class label information, and therefore to enhance the discriminability of subspace. Our approach is formulated as a constrained rank-minimization problem. We design an inexact augmented Lagrange multiplier optimization algorithm to solve it. Unlike the existing sparse representation and low-rank learning methods, our approach learns a low-dimensional subspace from recovered data, and explicitly incorporates the supervised information. Our approach and some baselines are evaluated on the COIL-100, ALOI, Extended YaleB, FERET, AR, and KinFace databases. The exper- imental results demonstrate the effectiveness of our approach, especially when the data contain considerable noise or variations.
Description
IEEE Transactions on Neural Networks and Learning Systems
The article of record as published may be found at http://dx.doi.org/10.1109/tnnls.2015.2464090
Rights
Copyright is reserved by the copyright owner.Collections
Related items
Showing items related by title, author, creator and subject.
-
Determining New System Design Requirements to Optimize Fleet Level Metrics under Uncertainty
Crossley, William (Monterey, California. Naval Postgraduate School, 2017-07); PUR-AM-17-208Traditional approaches to design and optimize a new system often do not consider how the operator will use this new system alongside the other existing systems. This モhand-offヤ between the designs of the new system and how ... -
IDENTIFICATION AND CLASSIFICATION OF SIGNALS USING GENERATIVE ADVERSARIAL NETWORKS
Ellison, Bart D. (Monterey, CA; Naval Postgraduate School, 2021-03);Research has shown that machine learning holds promise as a technique to improve the identification and classification of signals of interest. This study proposes the use of machine learning, specifically generative ... -
MACHINE LEARNING ANALYSIS OF THE ALCOHOL AND DRUG MANAGEMENT INFORMATION TRACKING SYSTEM
Angeles, Vanny Mae O. (Monterey, CA; Naval Postgraduate School, 2021-09);Substance abuse remains a significant problem among Sailors within the U.S. Excessive use of alcohol and drugs can be detrimental to a Sailor's health, safety, and naval service. The U.S. Navy recognizes that substance ...