Title :
On vector L0 penalized multivariate regression
Author :
Seneviratne, Akila J. ; Solo, Victor
Author_Institution :
Sch. of Electr. Eng. & Telecommun., Univ. of New South Wales, Sydney, NSW, Australia
Abstract :
The scalar sparse under-determined linear regression problem has had a rapid development with the multivariate version being of more recent interest. In this paper we pose a vector l0 penalized multivariate regression problem to generate coefficient vectors with shared sparsity profile and then solve the problem with a new cyclic descent algorithm. We give optimality conditions and also discuss penalty parameter selection. Finally we present simulation results that compare our algorithm with alternatives.
Keywords :
regression analysis; signal representation; cyclic descent algorithm; penalty parameter selection; signal representation; sparse under-determined linear regression problem; vector l0 penalized multivariate regression; Algorithm design and analysis; Approximation algorithms; Dictionaries; Matching pursuit algorithms; Multivariate regression; Signal to noise ratio; Vectors; cyclic descent; lo; multiple measurement vectors; multivariate regression; sparsity;
Conference_Titel :
Acoustics, Speech and Signal Processing (ICASSP), 2012 IEEE International Conference on
Conference_Location :
Kyoto
Print_ISBN :
978-1-4673-0045-2
Electronic_ISBN :
1520-6149
DOI :
10.1109/ICASSP.2012.6288698
