Online Hyperparameter-Free Sparse Estimation Method

Author

Zachariah, Dave ; Stoica, Petre

Author_Institution

Dept. of Inf. Technol., Uppsala Univ., Uppsala, Sweden

Volume

63

Issue

13

fYear

2015

fDate

1-Jul-15

Firstpage

3348

Lastpage

3359

Abstract

In this paper, we derive an online estimator for sparse parameter vectors which, unlike the LASSO approach, does not require the tuning of any hyperparameters. The algorithm is based on a covariance matching approach and is equivalent to a weighted version of the square-root LASSO. The computational complexity of the estimator is of the same order as that of the online versions of regularized least-squares (RLS) and LASSO. We provide a numerical comparison with feasible and infeasible implementations of the LASSO and RLS to illustrate the advantage of the proposed online hyperparameter-free estimator.

Keywords

compressed sensing; computational complexity; least squares approximations; parameter estimation; regression analysis; LASSO; RLS; computational complexity; covariance matching approach; online hyperparameter-free sparse estimation method; regularized least squares method; square-root LASSO weighted version; Complexity theory; Cost function; Covariance matrices; Estimation; Minimization; Noise; SPICE; Parameter estimation; recursive estimation; sparse parameters;

fLanguage

English

Journal_Title

Signal Processing, IEEE Transactions on

Publisher

ieee

ISSN

1053-587X

Type

jour

DOI

10.1109/TSP.2015.2421472

Filename

7084199