Minimax rates of estimation for high-dimensional linear regression over $\ell_q$-balls

TL;DR

This paper characterizes the minimax rates of estimation for high-dimensional linear regression over -balls, establishing bounds for -losses and prediction error under regularity conditions on the design matrix.

Contribution

It provides the first comprehensive analysis of minimax rates for -regularized linear regression, including lower bounds and constructive upper bounds for -norms and prediction loss.

Findings

Minimax error scales as (rac{\, ext{log}\, p}{n})^{1-rac{}{2}}.

Lower bounds are derived using information-theoretic methods.

Least-squares over -balls achieves minimax rates, with =0 case compared to -relaxations.

Abstract

Consider the standard linear regression model $\y = \Xmat \betastar + w$, where $\y \in \real^\numobs$ is an observation vector, $\Xmat \in \real^{\numobs \times \pdim}$ is a design matrix, $\betastar \in \real^\pdim$ is the unknown regression vector, and $w \sim \mathcal{N}(0, \sigma^2 I)$ is additive Gaussian noise. This paper studies the minimax rates of convergence for estimation of $\betastar$ for $\ell_\rpar$-losses and in the $\ell_2$-prediction loss, assuming that $\betastar$ belongs to an $\ell_{\qpar}$-ball $\Ballq(\myrad)$ for some $\qpar \in [0,1]$. We show that under suitable regularity conditions on the design matrix $\Xmat$, the minimax error in $\ell_2$-loss and $\ell_2$-prediction loss scales as $\Rq \big(\frac{\log \pdim}{n}\big)^{1-\frac{\qpar}{2}}$. In addition, we provide lower bounds on minimax risks in $\ell_{\rpar}$-norms, for all $\rpar \in [1, +\infty], \rpar \neq \qpar$. Our proofs of the lower bounds are information-theoretic in nature, based on Fano's inequality and results on the metric entropy of the balls $\Ballq(\myrad)$, whereas our proofs of the upper bounds are direct and constructive, involving direct analysis of least-squares over $\ell_{\qpar}$-balls. For the special case $q = 0$, a comparison with $\ell_2$-risks achieved by computationally efficient $\ell_1$-relaxations reveals that although such methods can achieve the minimax rates up to constant factors, they require slightly stronger assumptions on the design matrix $\Xmat$ than algorithms involving least-squares over the $\ell_0$-ball.