Subspaces of maximal operator spaces (Q1423742): Difference between revisions

From MaRDI portal
Import240304020342 (talk | contribs)
Set profile property.
Set OpenAlex properties.
Property / full work available at URL
 
Property / full work available at URL: https://doi.org/10.1007/s00020-002-1177-9 / rank
 
Normal rank
Property / OpenAlex ID
 
Property / OpenAlex ID: W2049550783 / rank
 
Normal rank

Revision as of 19:24, 19 March 2024

scientific article
Language Label Description Also known as
English
Subspaces of maximal operator spaces
scientific article

    Statements

    Subspaces of maximal operator spaces (English)
    0 references
    0 references
    7 March 2004
    0 references
    Recall that an operator space \(X\) is maximal if for any arbitrary operator space \(Y\), every bounded linear map \(u: X \to Y\) is automatically completely bounded with \(\| u\| _{cb}=\| u\| \). For any given Banach space \(X\), there is a unique maximal operator space whose underlying Banach space is \(X\). This operator space is denoted by \(\text{Max}(X)\) and the matrix norm on \(X\) is defined by \[ \left\| x\right\| _{\text{Max}(X) \otimes_{\min} B(H)}:= \sup\left\{ \left\| (u \otimes {Id}_{B(H)})x\right\| \;:\;u:X \to B(K), \| u\| \leq 1\right\}. \] The main purpose of the paper under review is to investigate structural aspects of subspaces of maximal operator spaces (called submaximal spaces). The main characterization can be summarized as follows: An operator space \(X\) is submaximal if and only if there is a Hilbert space \(H\) such that there exists a complete isometry from \(X\) into \(\text{Max}(B(H))\). Various results on complete isomorphism versions of the above characterization are also included. The author then considers the space of \(n\)-dimensional submaximal operator spaces (denoted by \(\mathcal{SM}_n)\) equipped with the \(cb\)-Banach-Mazur distance. The main result along this line is that \(\mathcal{SM}_n\) is closed but not compact. The last part of the paper deals with concrete cases, namely, operator subspaces of \(\text{Max}(L_1)\) and operator subspaces of \(\text{Max}(L_\infty)\). For the case of \(\text{Max}(L_1)\), the author focuses on Hilbertian subspaces. The main result for this special case is the following: If \(E\) is a homogeneous Hilbertian subspace of \(\text{Max}(L_1)\), then \(E\) is \(cb\)-isomorphic to \(R+C\) where \(R\) and \(C\) denote the row and column Hilbert spaces, respectively. For the case of \(\text{Max}(L_\infty)\), the author introduces a ``smallest'' submaximal operator space structure on any given Banach space. Let \(E\) be a Banach space and \(J\) an isometric embedding of \(E\) into a commutative \(C^*\)-algebra \(A\). Denote by \(\mu(E)\) the space \(J(E)\) with operator space structure inherited form \(\text{Max}(A)\) (it is verified that \(\mu(E)\) does not depend on \(J\) and \(A\)). The author provides estimates of the \(cb\)-distance between \(\mu(l^n_2)\) and some classical Hilbertian operator spaces such as \(\text{Min}(l^n_2)\), \(\text{Max}(l^n_2)\), \(C_n\), \(R_n\), \(OH_n\), among others.
    0 references
    operator spaces
    0 references
    submaximal
    0 references

    Identifiers

    0 references
    0 references
    0 references
    0 references
    0 references
    0 references