Skip to content
DraconiticResearch
Search Sources API
201,642 studies31 sourcesUpdated 11 min ago
PreprintarXiv licencearXiv

A Truncated Singular-Value Bound for Spectral Variation of Normal Matrices

Abstract

For normal matrices A and B, the classical Hoffman-Wielandt theorem bounds the optimal matching distance between their spectra by the Frobenius norm of A-B. We prove the sharper estimate d(σ(A),σ(B))^2 \le \sum_{k=1}^{\lfloor (n+1)/2 \rfloor} s_k(A-B)^2, where s_k(M) are the singular values of M, involving only the first \lfloor (n+1)/2 \rfloor singular values of the difference. Consequently, d(σ(A),σ(B)) \le \sqrt{\lfloor (n+1)/2 \rfloor}\,\|A-B\|, which improves the classical dimension-free bound for all 3 \le n \le 16. The proof combines a min-max duality for optimal matching distances with spectral subspace overlaps and the monotonicity of singular values under rectangular compressions.

Related

  1. Spectral density estimation for normal matrices
    Cameron Musco, Christopher Musco, Rikhav Shah +2
    arXivPreprintarXiv licenceFull textNumerical analysis & scientific computingarXiv
  2. Weak type (1,1) estimates for noncommutative maximal truncated rough singular integrals
    Junming Cao, Wei Li
    arXivPreprintarXiv licenceFull textAnalysis & PDEsarXiv
  3. Weak type (1,1) bound criterion for maximal truncated singular integral operators with rough kernels
    Xudong Lai, Lingxin Xu
    arXivPreprintarXiv licenceFull textAnalysis & PDEsarXiv
  4. Semialgebraic Dimension and Truncated Toeplitz Models for Complex Symmetric Matrices
    Ryan O'Loughlin
    arXivPreprintCC0Full textAnalysis & PDEsarXiv
  5. Least singular values of Gaussian matrices with deterministic shifts
    Xuanang Hu, Zeyan Song, Xinglong Wu
    arXivPreprintarXiv licenceFull textProbabilityarXiv
BibTeX
@misc{tang2026truncated,
  title = {{A Truncated Singular-Value Bound for Spectral Variation of Normal Matrices}},
  author = {Qiyue Tang},
  howpublished = {arXiv},
  year = {2026},
  month = oct,
  doi = {10.48550/arxiv.2609.09177},
  eprint = {2609.09177},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2609.09177},
}
RIS
TY  - UNPB
TI  - A Truncated Singular-Value Bound for Spectral Variation of Normal Matrices
AU  - Qiyue Tang
PY  - 2026
DA  - 2026/10/02
PB  - arXiv
DO  - 10.48550/arxiv.2609.09177
UR  - https://arxiv.org/abs/2609.09177
AB  - For normal matrices A and B, the classical Hoffman-Wielandt theorem bounds the optimal matching distance between their spectra by the Frobenius norm of A-B. We prove the sharper estimate d(σ(A),σ(B))^2 \le \sum_{k=1}^{\lfloor (n+1)/2 \rfloor} s_k(A-B)^2, where s_k(M) are the singular values of M, involving only the first \lfloor (n+1)/2 \rfloor singular values of the difference. Consequently, d(σ(A),σ(B)) \le \sqrt{\lfloor (n+1)/2 \rfloor}\,\|A-B\|, which improves the classical dimension-free bound for all 3 \le n \le 16. The proof combines a min-max duality for optimal matching distances with spectral subspace overlaps and the monotonicity of singular values under rectangular compressions.
ER  -