Well-Conditioned Oblivious Perturbations in Linear Space

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Chenakkod, Shabarish, Dereziński, Michał, Dong, Xiaoyu, Rudelson, Mark
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915958131523584
author Chenakkod, Shabarish
Dereziński, Michał
Dong, Xiaoyu
Rudelson, Mark
author_facet Chenakkod, Shabarish
Dereziński, Michał
Dong, Xiaoyu
Rudelson, Mark
contents Perturbing a deterministic $n$-dimensional matrix with small Gaussian noise is a cornerstone of smoothed analysis of algorithms [Spielman and Teng, JACM 2004], as it reduces the condition number of the input to $O(n)$, and with it the complexity of many matrix algorithms. However, when deployed algorithmically, these perturbations are expensive due to the cost of generating and storing $n^2$ Gaussian random variables. We propose a perturbation that requires generating and storing $O(n)$ random numbers in $O(\log n)$ bits of precision, and reduces the condition number of any deterministic matrix to $O(n)$, matching Gaussian perturbations. Our result in particular implies a better complexity for the perturbed conjugate gradient algorithm, showing that we can solve an $n\times n$ linear system in linear space to within an arbitrarily small constant backward error using $O(n)$ matrix-vector products. In our construction, we introduce the concept of a pattern matrix, which is a dense deterministic matrix that maps all sparse vectors into dense vectors, and we combine it with a sparse perturbation whose entries are dependent and located in a non-uniform fashion. In order to analyze this construction, we develop new techniques for lower bounding the smallest singular value of a random matrix with dependent entries.
format Preprint
id arxiv_https___arxiv_org_abs_2604_23193
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Well-Conditioned Oblivious Perturbations in Linear Space
Chenakkod, Shabarish
Dereziński, Michał
Dong, Xiaoyu
Rudelson, Mark
Data Structures and Algorithms
Machine Learning
Numerical Analysis
Probability
Perturbing a deterministic $n$-dimensional matrix with small Gaussian noise is a cornerstone of smoothed analysis of algorithms [Spielman and Teng, JACM 2004], as it reduces the condition number of the input to $O(n)$, and with it the complexity of many matrix algorithms. However, when deployed algorithmically, these perturbations are expensive due to the cost of generating and storing $n^2$ Gaussian random variables. We propose a perturbation that requires generating and storing $O(n)$ random numbers in $O(\log n)$ bits of precision, and reduces the condition number of any deterministic matrix to $O(n)$, matching Gaussian perturbations. Our result in particular implies a better complexity for the perturbed conjugate gradient algorithm, showing that we can solve an $n\times n$ linear system in linear space to within an arbitrarily small constant backward error using $O(n)$ matrix-vector products. In our construction, we introduce the concept of a pattern matrix, which is a dense deterministic matrix that maps all sparse vectors into dense vectors, and we combine it with a sparse perturbation whose entries are dependent and located in a non-uniform fashion. In order to analyze this construction, we develop new techniques for lower bounding the smallest singular value of a random matrix with dependent entries.
title Well-Conditioned Oblivious Perturbations in Linear Space
topic Data Structures and Algorithms
Machine Learning
Numerical Analysis
Probability
url https://arxiv.org/abs/2604.23193