Sparse Recovery With Unknown Variance: A LASSO-Type Approach

DOI10.1109/TIT.2014.2301162zbMATH Open1360.62396arXiv1101.0434OpenAlexW2165186963MaRDI QIDQ2986269FDOQ2986269

Authors: Stéphane Chrétien, Sébastien Darses 

Publication date: 16 May 2017

Published in: IEEE Transactions on Information Theory (Search for Journal in Brave)

Abstract: We address the issue of estimating the regression vector $$ in the generic $s$-sparse linear model $$, with $$, $yin{R}^n$, $zsimmathcalN(0,s{g}^{2}I)$ and $p>n$ when the variance $s{g}^2$ is unknown. We study two LASSO-type methods that jointly estimate $$ and the variance. These estimators are minimizers of the $el{l}_1$ penalized least-squares functional, where the relaxation parameter is tuned according to two different strategies. In the first strategy, the relaxation parameter is of the order $chsigmasqrtlogp$, where $ch{sigma}^2$ is the empirical variance. %The resulting optimization problem can be solved by running only a few successive LASSO instances with %recursive updating of the relaxation parameter. In the second strategy, the relaxation parameter is chosen so as to enforce a trade-off between the fidelity and the penalty terms at optimality. For both estimators, our assumptions are similar to the ones proposed by Cand`es and Plan in {it Ann. Stat. (2009)}, for the case where $s{g}^2$ is known. We prove that our estimators ensure exact recovery of the support and sign pattern of $$ with high probability. We present simulations results showing that the first estimator enjoys nearly the same performances in practice as the standard LASSO (known variance case) for a wide range of the signal to noise ratio. Our second estimator is shown to outperform both in terms of false detection, when the signal to noise ratio is low.

Full work available at URL: https://arxiv.org/abs/1101.0434

Recommendations

• Sparse recovery by reduced variance stochastic approximation
• Lasso-type recovery of sparse representations for high-dimensional data
• On the sparsity of Lasso minimizers in sparse data recovery
• Joint sparse recovery based on variances
• The sparsity of LASSO-type minimizers
• Sparse recovery under matrix uncertainty
• Sparse recovery under weak moment assumptions
• A unified approach to model selection and sparse recovery using regularized least squares
• Estimation of covariance matrix via the sparse Cholesky factor with lasso
• Sparse Recovery via Partial Regularization: Models, Theory, and Algorithms

Mathematics Subject Classification ID

Ridge regression; shrinkage estimators (Lasso) (62J07)

Cited In (4)

• Reducing effects of bad data using variance based joint sparsity recovery
• A fast algorithm for the semi-definite relaxation of the state estimation problem in power grids
• Decomposition of dynamical signals into jumps, oscillatory patterns, and possible outliers
• \(\ell_1\)-penalised ordinal polytomous regression estimators with application to gene expression studies

This page was built for publication: Sparse Recovery With Unknown Variance: A LASSO-Type Approach

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2986269)