Numerical Linear Algebra in Linear Space

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Liu, Yiping, Nguyen, Hoai-An, Yang, Junzhao
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909674696081408
author Liu, Yiping
Nguyen, Hoai-An
Yang, Junzhao
author_facet Liu, Yiping
Nguyen, Hoai-An
Yang, Junzhao
contents We present a randomized linear-space solver for general linear systems $\mathbf{A} \mathbf{x} = \mathbf{b}$ with $\mathbf{A} \in \mathbb{Z}^{n \times n}$ and $\mathbf{b} \in \mathbb{Z}^n$, without any assumption on the condition number of $\mathbf{A}$. For matrices whose entries are bounded by $\mathrm{poly}(n)$, the solver returns a $(1+ε)$-multiplicative entry-wise approximation to vector $\mathbf{x} \in \mathbb{Q}^{n}$ using $\widetilde{O}(n^2 \cdot \mathrm{nnz}(\mathbf{A}))$ bit operations and $O(n \log n)$ bits of working space (i.e., linear in the size of a vector), where $\mathrm{nnz}$ denotes the number of nonzero entries. Our solver works for right-hand vector $\mathbf{b}$ with entries up to $n^{O(n)}$. To our knowledge, this is the first linear-space linear system solver over the rationals that runs in $\widetilde{O}(n^2 \cdot \mathrm{nnz}(\mathbf{A}))$ time. We also present several applications of our solver to numerical linear algebra problems, for which we provide algorithms with efficient polynomial running time and near-linear space. In particular, we present results for linear regression, linear programming, eigenvalues and eigenvectors, and singular value decomposition.
format Preprint
id arxiv_https___arxiv_org_abs_2507_02433
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Numerical Linear Algebra in Linear Space
Liu, Yiping
Nguyen, Hoai-An
Yang, Junzhao
Data Structures and Algorithms
We present a randomized linear-space solver for general linear systems $\mathbf{A} \mathbf{x} = \mathbf{b}$ with $\mathbf{A} \in \mathbb{Z}^{n \times n}$ and $\mathbf{b} \in \mathbb{Z}^n$, without any assumption on the condition number of $\mathbf{A}$. For matrices whose entries are bounded by $\mathrm{poly}(n)$, the solver returns a $(1+ε)$-multiplicative entry-wise approximation to vector $\mathbf{x} \in \mathbb{Q}^{n}$ using $\widetilde{O}(n^2 \cdot \mathrm{nnz}(\mathbf{A}))$ bit operations and $O(n \log n)$ bits of working space (i.e., linear in the size of a vector), where $\mathrm{nnz}$ denotes the number of nonzero entries. Our solver works for right-hand vector $\mathbf{b}$ with entries up to $n^{O(n)}$. To our knowledge, this is the first linear-space linear system solver over the rationals that runs in $\widetilde{O}(n^2 \cdot \mathrm{nnz}(\mathbf{A}))$ time. We also present several applications of our solver to numerical linear algebra problems, for which we provide algorithms with efficient polynomial running time and near-linear space. In particular, we present results for linear regression, linear programming, eigenvalues and eigenvectors, and singular value decomposition.
title Numerical Linear Algebra in Linear Space
topic Data Structures and Algorithms
url https://arxiv.org/abs/2507.02433