Universality of regularized regression estimators in high dimensions

TL;DR

This paper develops a universality framework for regularized regression estimators in high dimensions, extending Gaussian-based analysis to broader design matrices and validating inference procedures.

Contribution

It introduces a structural universality framework compatible with CGMT, allowing high-dimensional analysis of estimators beyond Gaussian designs.

Findings

Universality properties hold for Ridge, Lasso, and robust regression estimators under general designs.

Validated inference procedures for Lasso using degrees-of-freedom adjustment in non-Gaussian settings.

Counterexample shows universality does not extend to all isotropic designs.

Abstract

The Convex Gaussian Min-Max Theorem (CGMT) has emerged as a prominent theoretical tool for analyzing the precise stochastic behavior of various statistical estimators in the so-called high dimensional proportional regime, where the sample size and the signal dimension are of the same order. However, a well recognized limitation of the existing CGMT machinery rests in its stringent requirement on the exact Gaussianity of the design matrix, therefore rendering the obtained precise high dimensional asymptotics largely a specific Gaussian theory in various important statistical models. This paper provides a structural universality framework for a broad class of regularized regression estimators that is particularly compatible with the CGMT machinery. In particular, we show that with a good enough $\ell_\infty$ bound for the regression estimator $\hat{\mu}_A$, any `structural property' that can be detected via the CGMT for $\hat{\mu}_G$ (under a standard Gaussian design $G$) also holds for $\hat{\mu}_A$ under a general design $A$ with independent entries. As a proof of concept, we demonstrate our new universality framework in three key examples of regularized regression estimators: the Ridge, Lasso and regularized robust regression estimators, where new universality properties of risk asymptotics and/or distributions of regression estimators and other related quantities are proved. As a major statistical implication of the Lasso universality results, we validate inference procedures using the degrees-of-freedom adjusted debiased Lasso under general design and error distributions. We also provide a counterexample, showing that universality properties for regularized regression estimators do not extend to general isotropic designs.