Restricted eigenvalue property for corrupted Gaussian designs

TL;DR

This paper establishes conditions under which Gaussian designs with corrupted samples satisfy a robust Restricted Eigenvalue property, enabling high-dimensional inference despite outliers without relying on traditional bounds.

Contribution

It introduces a new technique for proving Restricted Eigenvalue conditions in corrupted Gaussian designs that does not depend on extreme singular value bounds or mutual incoherence.

Findings

Sharper restricted eigenvalue constants are obtained.

Sparsity and outlier count can be independent.

Conditions are suitable for robust high-dimensional inference.

Abstract

Motivated by the construction of tractable robust estimators via convex relaxations, we present conditions on the sample size which guarantee an augmented notion of Restricted Eigenvalue-type condition for Gaussian designs. Such a notion is suitable for high-dimensional robust inference in a Gaussian linear model and a multivariate Gaussian model when samples are corrupted by outliers either in the response variable or in the design matrix. Our proof technique relies on simultaneous lower and upper bounds of two random bilinear forms with very different behaviors. Such simultaneous bounds are used for balancing the interaction between the parameter vector and the estimated corruption vector as well as for controlling the presence of corruption in the design. Our technique has the advantage of not relying on known bounds of the extreme singular values of the associated Gaussian ensemble nor on the use of mutual incoherence arguments. A relevant consequence of our analysis, compared to prior work, is that a significantly sharper restricted eigenvalue constant can be obtained under weaker assumptions. In particular, the sparsity of the unknown parameter and the number of outliers are allowed to be completely independent of each other.