DocumentCode :
1765886
Title :
Nonparametric Basis Pursuit via Sparse Kernel-Based Learning: A Unifying View with Advances in Blind Methods
Author :
Bazerque, Juan Andres ; Giannakis, Georgios
Author_Institution :
Dept. of ECE & Digital Technol. Center, Univ. of Minnesota, Minneapolis, MN, USA
Volume :
30
Issue :
4
fYear :
2013
fDate :
41456
Firstpage :
112
Lastpage :
125
Abstract :
Signal processing tasks as fundamental as sampling, reconstruction, minimum mean-square error interpolation, and prediction can be viewed under the prism of reproducing kernel Hilbert spaces (RKHSs). Endowing this vantage point with contemporary advances in sparsity-aware modeling and processing promotes the nonparametric basis pursuit advocated in this article as the overarching framework for the confluence of kernel-based learning (KBL) approaches leveraging sparse linear regression, nuclear-norm regularization, and dictionary learning. The novel sparse KBL toolbox goes beyond translating sparse parametric approaches to their nonparametric counterparts to incorporate new possibilities such as multikernel selection and matrix smoothing. The impact of sparse KBL to signal processing applications is illustrated through test cases from cognitive radio sensing, microarray data imputation, and network traffic prediction.
Keywords :
Hilbert spaces; cognitive radio; learning (artificial intelligence); matrix algebra; mean square error methods; regression analysis; signal sampling; KBL toolbox; RKHS; blind methods; cognitive radio sensing; dictionary learning; kernel Hilbert spaces; microarray data imputation; minimum mean square error interpolation; network traffic prediction; nonparametric basis pursuit; nuclear-norm regularization; signal processing; signal reconstruction; signal sampling; sparse kernel-based learning; sparse linear regression; sparse parametric approach translation; sparsity-aware modelling; Kernel; Learning systems; Machine learning; Signal processing algorithms; Sparse matrices; Statistical analysis;
fLanguage :
English
Journal_Title :
Signal Processing Magazine, IEEE
Publisher :
ieee
ISSN :
1053-5888
Type :
jour
DOI :
10.1109/MSP.2013.2253354
Filename :
6530741
Link To Document :
بازگشت