A Hilbert-Schmidt norm inequality for positive semidefinite matrices related to a question of Bourin (Q2154920)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | A Hilbert-Schmidt norm inequality for positive semidefinite matrices related to a question of Bourin |
scientific article |
Statements
A Hilbert-Schmidt norm inequality for positive semidefinite matrices related to a question of Bourin (English)
0 references
15 July 2022
0 references
Let \(A\) and \(B\) be positive semidefinite matrices of the same size. The main result of the paper is the inequality \[ \| B^{1-t}A^{2t-1}B^{1-t} + A^{1-t}B^{2t-1}A^{1-t} \|_F \leq \| A + B \|_F \] for \(1/2 \leq t \leq 3/4\), where \(\| X \|_F = \sqrt{\mathrm{tr}(X^*X)} = \sqrt{\sum_{i,j} |x_{ij}|^2}\) is the Frobenius norm of the matrix \(X\) (in the paper, the authors use the name Hilbert-Schmidt norm). As a corollary the authors prove that \[ \|A^t B^{1-t} + B^t A^{1-t} \|_F \leq \| A + B \|_F \] for \(1/4 \leq t \leq 3/4\).
0 references
trace
0 references
Hilbert-Schmidt norm
0 references
positive semidefinite matrix
0 references
Bourin's question
0 references
inequality
0 references
0 references
0 references