Analysis of perturbations of singular values in concatenated matrices

UDC 512.5 Concatenating matrices is a common technique for uncovering shared structures in the data by using singular-valued decomposition (SVD) and low-rank approximations. The main analyzed question is to determine the relationship between the singular-valued spectrum of the concatenated matrix an...

Full description

Saved in:
Bibliographic Details
Date:2026
Main Authors: Shamrai, M., Шамрай, Максим
Format: Article
Language:Ukrainian
Published: Institute of Mathematics, NAS of Ukraine 2026
Online Access:https://umj.imath.kiev.ua/index.php/umj/article/view/9103
Tags: Add Tag
No Tags, Be the first to tag this record!
Journal Title:Ukrains’kyi Matematychnyi Zhurnal
Download file: Pdf

Institution

Ukrains’kyi Matematychnyi Zhurnal
_version_ 1860513387983667200
author Shamrai, M.
Шамрай, Максим
author_facet Shamrai, M.
Шамрай, Максим
author_sort Shamrai, M.
baseUrl_str https://umj.imath.kiev.ua/index.php/umj/oai
collection OJS
datestamp_date 2026-03-22T13:31:15Z
description UDC 512.5 Concatenating matrices is a common technique for uncovering shared structures in the data by using singular-valued decomposition (SVD) and low-rank approximations. The main analyzed question is to determine the relationship between the singular-valued spectrum of the concatenated matrix and  the spectra of its individual components. In the present work, we develop a perturbation technique that extends the classical results, such as Weyl's inequality, to the case of concatenated matrices. We establish the analytic bounds, which give quantitative characteristics of the stability of singular values under small perturbations in submatrices. The obtained results demonstrate that if the submatrices are close in a certain norm, then the predominant singular values of the concatenated matrix remain stable, which enables us to control the compromise between the accuracy and compression. The accumulated results also lay the theoretical basis for the improved matrix clustering and compression strategies with applications in the numerical linear algebra, signal processing, and data-driven modeling.
doi_str_mv 10.3842/umzh.v77i8.9103
first_indexed 2026-03-24T03:43:53Z
format Article
fulltext Skip to main content Skip to main navigation menu Skip to site footer Open Menu Ukrains’kyi Matematychnyi Zhurnal Current Archives Submissions Major topics of interest About About Journal Editorial Team Ethics & Disclosures Contacts Search Register Login Home / Login Login Required fields are marked with an asterisk: * Subscription required to access item. To verify subscription, log in to journal. Login Username or Email * Required Password * Required Forgot your password? Keep me logged in Login Register Language English Українська Information For Readers For Authors For Librarians subscribe Subscribe Latest publications Make a Submission Make a Submission STM88 menghadirkan Link Gacor dengan RTP tinggi untuk peluang menang yang lebih sering! Bergabunglah sekarang dan buktikan keberuntungan Anda! Pilih STM88 sebagai agen toto terpercaya Anda dan nikmati kenyamanan bermain dengan sistem betting cepat, result resmi, dan bonus cashback harian.
id umjimathkievua-article-9103
institution Ukrains’kyi Matematychnyi Zhurnal
keywords_txt_mv keywords
language Ukrainian
last_indexed 2026-03-24T03:43:53Z
publishDate 2026
publisher Institute of Mathematics, NAS of Ukraine
record_format ojs
resource_txt_mv umjimathkievua/0d/edf8ab9c653c70ce74cbca61a53c680d
spelling umjimathkievua-article-91032026-03-22T13:31:15Z Analysis of perturbations of singular values in concatenated matrices Аналіз збурень сингулярних значень конкатенованих матриць Shamrai, M. Шамрай, Максим numerical analysis perturbation analysis singular values concatenated matrices clustering Singular Value Decomposition matrix theory Numerical Analysis Perturbation Analysis Matrix theory UDC 512.5 Concatenating matrices is a common technique for uncovering shared structures in the data by using singular-valued decomposition (SVD) and low-rank approximations. The main analyzed question is to determine the relationship between the singular-valued spectrum of the concatenated matrix and  the spectra of its individual components. In the present work, we develop a perturbation technique that extends the classical results, such as Weyl's inequality, to the case of concatenated matrices. We establish the analytic bounds, which give quantitative characteristics of the stability of singular values under small perturbations in submatrices. The obtained results demonstrate that if the submatrices are close in a certain norm, then the predominant singular values of the concatenated matrix remain stable, which enables us to control the compromise between the accuracy and compression. The accumulated results also lay the theoretical basis for the improved matrix clustering and compression strategies with applications in the numerical linear algebra, signal processing, and data-driven modeling. УДК 512.5 Конкатенація матриць є поширеним прийомом для виявлення спільної структури в даних за допомогою сингулярного розкладу матриць (SVD) та низькорангової апроксимації. Основне питання полягає в тому як спектр сингулярних значень конкатенованої матриці пов’язаний зі спектрами її складових. У цій статті розширено техніку збурень, що узагальнює класичні результати, такі як нерівність Вейля, на випадок конкатенованих матриць. Встановлено аналітичні межі, які кількісно характеризують стабільність сингулярних значень при малих збуреннях її підматриць. Показано, що якщо підматриці близькі за певною нормою, домінантні сингулярні значення конкатенованої матриці залишаються стабільними, що дає змогу керувати компромісом між точністю та стисненням. Отримані результати забезпечують теоретичну основу для вдосконалених стратегій кластеризації та стиснення матриць із застосуваннями в чисельній лінійній алгебрі, обробці сигналів та моделях, побудованих на даних. Institute of Mathematics, NAS of Ukraine 2026-03-21 Article Article application/pdf https://umj.imath.kiev.ua/index.php/umj/article/view/9103 10.3842/umzh.v77i8.9103 Ukrains’kyi Matematychnyi Zhurnal; Vol. 77 No. 8 (2025); 521–532 Український математичний журнал; Том 77 № 8 (2025); 521–532 1027-3190 uk https://umj.imath.kiev.ua/index.php/umj/article/view/9103/10557 Copyright (c) 2025 Максим Шамрай
spellingShingle Shamrai, M.
Шамрай, Максим
Analysis of perturbations of singular values in concatenated matrices
title Analysis of perturbations of singular values in concatenated matrices
title_alt Аналіз збурень сингулярних значень конкатенованих матриць
title_full Analysis of perturbations of singular values in concatenated matrices
title_fullStr Analysis of perturbations of singular values in concatenated matrices
title_full_unstemmed Analysis of perturbations of singular values in concatenated matrices
title_short Analysis of perturbations of singular values in concatenated matrices
title_sort analysis of perturbations of singular values in concatenated matrices
topic_facet numerical analysis
perturbation analysis
singular values
concatenated matrices
clustering
Singular Value Decomposition
matrix theory
Numerical Analysis
Perturbation Analysis
Matrix theory
url https://umj.imath.kiev.ua/index.php/umj/article/view/9103
work_keys_str_mv AT shamraim analysisofperturbationsofsingularvaluesinconcatenatedmatrices
AT šamrajmaksim analysisofperturbationsofsingularvaluesinconcatenatedmatrices
AT shamraim analízzburenʹsingulârnihznačenʹkonkatenovanihmatricʹ
AT šamrajmaksim analízzburenʹsingulârnihznačenʹkonkatenovanihmatricʹ