Norm bounds for operator extensions
Consider the partial \(n\times n\) matrix \(F\) with bounded Hilbert space operator entries, where the lower triangular entries are specified and the strictly upper triangular entries are to be determined. Any choice of the strictly upper triangular entries provides a completion or extension of \(F\). The approach of this paper is to reduce the analysis of \(n\times n\) case to iterations of \(2\times 2\) completions. Let \(0<\nu, \mu\leq \delta\) and \(\delta> \min\{ \nu,\mu \}\). The operator matrix \({{b\;x} \choose {a\;c}}\) is called a \((\nu, \mu, \delta)\)-central completion if \(|{b\choose a}|\leq\nu\), \(|(a, c)|\leq \mu\) and \(x=- b(\delta^2 I-a^* a)^{-1} a^* c\). The authors characterize such central completions and analyze the upper bound for such central completions. If \(F\) is an \(n\times n\) operator matrix, then its central completion is obtained by determining as above the central completion of the first two columns of \(F\), then of the first three columns, and so on. They prove that the upper bounds are attained for \(2\times 2\) case and cannot be attained for \(n\times n\) cases with \(n>2\).
- scientific article; zbMATH DE number 5835621 (Why is no real title available?)
- scientific article; zbMATH DE number 3881329 (Why is no real title available?)
- Bounds for operator/Hilbert-Schmidt norm minimization using entropy
- Exact bounds for the minimum norm of extension operators for Sobolev spaces
- Operator-valued Extensions of Matrix-norm Inequalities
This page was built for publication: Norm bounds for operator extensions
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q1910822)
