On the sum of superoptimal singular values

From MaRDI portal
Publication:2391257



Abstract: We discuss the following extremal problem and its relevance to the sum of the so-called superoptimal singular values of a matrix function: Given an mimesn matrix function Phi on the unit circle mathbbT, when is there a matrix function Psi∗ in the set Akn,m such that int_{mathbb{T}}{ m trace}(Phi(zeta)Psi_{*}(zeta))dm(zeta)=sup_{Psiin A_{k}^{n,m}}|int_{mathbb{T}}{ m trace}(Phi(zeta)Psi(zeta))dm(zeta)|? The set Akn,m is defined by A_{k}^{n,m}={Psiin H_{0}^{1}: |Psi|_{L^{1}}leq 1, { m rank}Psi(zeta)leq k{a.e.}zetain T}. We introduce Hankel-type operators on spaces of matrix functions and prove that this problem has a solution if and only if the corresponding Hankel-type operator has a maximizing vector. We also characterize the smallest number k for which int_{mathbb{T}}{ m trace}(Phi(zeta)Psi(zeta))dm(zeta) equals the sum of all the superoptimal singular values of an admissible matrix function Phi for some PsiinAkn,m. Moreover, we provide a representation of any such function Psi when Phi is an admissible very badly approximable unitary-valued nimesn matrix function.


The present paper is based on the author's PhD dissertation. Its aim is to give a characterization of the smallest number for which the sum of the so-called superoptimal singular values of an admissible matrix function is evaluated. The paper is organized in five sections, the first two of which containing the main definitions, notations, and some principles of functional analysis expressed in matrix language. The notions of superoptimal approximation and very badly approximation by bounded analytic matrix functions are explained. In particular, explicit formulations are given for the space \(L^p(S_q^{m,n})\), where \(S_q^{m,n}\) is the space of the \(m\times n\) matrices equipped with the Schatten-von Neumann norm. In the third section, Hankel-type operators on spaces of matrix functions are introduced, and the approximation is given in terms of Hankel-type operators. In the fourth section, the main result of the paper on the best approximation is obtained. The last section of the paper is concerned with the class of admissible unitary valued very badly approximable \(n\times n\) matrix functions.











This page was built for publication: On the sum of superoptimal singular values

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2391257)