The Pseudospectrum of Random Compressions of Matrices

TL;DR

This paper investigates the pseudospectrum of randomly compressed matrices, providing bounds on its expected area and establishing probabilistic estimates for singular values, with implications for stability analysis.

Contribution

It introduces new bounds on the pseudospectrum of random matrix compressions and develops advanced probabilistic estimates for singular values and quadratic forms.

Findings

Expected pseudospectrum area bounded by poly(n)log^2(1/ε)·ε^β

Tail bounds for least singular value of compressions

Non-asymptotic small-ball estimates for non-Hermitian quadratic forms

Abstract

The compression of a matrix $A\in\mathbb C^{n\times n}$ onto a subspace $V\subset\mathbb C^n$ is the matrix $Q^*AQ$ where the columns of $Q$ form an orthonormal basis for $V$. This is an important object in both operator theory and numerical linear algebra. Of particular interest are the eigenvalues of the compression and their stability under perturbations. This paper considers compressions onto subspaces sampled from the Haar measure on the complex Grassmannian. We show the expected area of the $\varepsilon$-pseudospectrum of such compressions is bounded by $\text{poly}(n)\log^2(1/\varepsilon)\cdot\varepsilon^\beta$, where $\beta=6/5,4/3$, or $2$ depending on some mild assumptions on $A$. Along the way, we obtain (a) tail bounds for the least singular value of compressions and (b) non-asymptotic small-ball estimates for random non-Hermitian quadratic forms surpassing bounds achieved by existing methods.