No label defined (Q4248263)
From MaRDI portal
!
This is the item page for this Wikibase entity, intended for internal use and editing purposes. Please use the normal view instead:
scientific article; zbMATH DE number 1296384
| Language | Label | Description | Also known as |
|---|---|---|---|
| default for all languages | No label defined |
||
| English | No label defined |
scientific article; zbMATH DE number 1296384 |
Statements
13 July 1999
0 references
eigenvalues
0 references
inequality
0 references
sum of operators
0 references
Schubert calculus
0 references
tensor products
0 references
0.8931257128715515
0 references
0.8767044544219971
0 references
0.8519229292869568
0 references
In the simplest form, the problem solved by \textit{A. A. Klyachko} [Stable bundles, representation theory and Hermitian operators, Institute Mittag-Leffler Preprint (1996-97)] is to describe the eigenvalues of the sum \(A+B\) of two Hermitian \(n{\times}n\) matrices in terms of the eigenvalues of \(A\) and \(B\). List the eigenvalues of a Hermitian matrix \(H\) in decreasing order \(\lambda(H): \lambda_1(H) \geq \lambda_2(H) \geq \cdots \geq \lambda_n(H)\). Write \(\lambda(A): \alpha_1 \geq \alpha_2 \geq \cdots \geq \alpha_n\), \(\lambda(B): \beta_1 \geq \beta_2 \geq \cdots \geq \beta_n\); let \(C = A + B\), and \(\lambda(C): \gamma_1 \geq \gamma_2 \geq \cdots \geq \gamma_n\) Of course one always has the trace identity \(\sum{\gamma_i} = \sum{\alpha_i} + \sum{\beta_i}\). The problem arose from questions in solid mechanics, where eigenvalues of symmetric matrices determine shapes of ellipsoids. The inequality \(\gamma_1 \leq \alpha_1 + \beta_1\) was known more then a century ago. For \(n = 2\) the inequalities \(\gamma_1 \leq \alpha_1 + \beta_1\), \(\gamma_2 \leq \alpha_1 + \beta_2\) and \(\gamma_2 \leq \alpha_2 + \beta_1\) are easily seen to be sufficient to characterize the possible eigenvalues of the sum. In this paper the author rephrases and proves some of Klyachko's theorems which generalize to sums of more than two Hermitian matrices, and to tensor products of more than two representations. His methods and discussions are based on the Schubert calculus. The final section of the paper contains some applications, and a brief discussion of some more recent works on eigenvalues of unitary matrices. For the entire collection see [Zbl 0911.00019].
0 references