The isometric extension of the into mapping from a \(\mathcal{L}^{\infty}(\Gamma)\)-type space to some Banach space (Q2382956)

From MaRDI portal
Revision as of 06:57, 5 March 2024 by Import240304020342 (talk | contribs) (Set profile property.)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
scientific article
Language Label Description Also known as
English
The isometric extension of the into mapping from a \(\mathcal{L}^{\infty}(\Gamma)\)-type space to some Banach space
scientific article

    Statements

    The isometric extension of the into mapping from a \(\mathcal{L}^{\infty}(\Gamma)\)-type space to some Banach space (English)
    0 references
    5 October 2007
    0 references
    Given two real Banach spaces \(E\) and \(F\), and a subset \(S\) of \(E\), a mapping \(V_0:S \rightarrow F\) is called an isometry if \(\| V_0(x_1)-V_0(x_2)\| =\| x_1 - x_2\| \) for all \(x_1,x_2\in S\). The classical Mazur-Ulam theorem states that every surjective isometry between two normed spaces is affine (i.e., a linear mapping up to translation). One of the possible ways to extend this result is the study of the isometric extension problem posed by \textit{D.\ Tingley} [Geom. Dedicata 22, No. 3, 371--378 (1987; Zbl 0615.51005)]: Let \(E\) and \(F\) be real normed spaces with unit spheres \(S(E)\) and \(S(F)\), respectively. Suppose that \(V_0:S(E)\rightarrow S(F)\) is a surjective isometry. Is then \(V_0\) necessarily the restriction of a linear isometry between \(E\) and \(F\)? The paper under review deals with this question when the first space is an \({\mathcal L}^\infty (\Gamma)\)-type space and the isometry between the unit spheres is not necessarily surjective. The main result of the paper is the following. Let \(E\) be a real Banach space, \(X=\ell_\infty(\Gamma)\), and let \(V_0:S(X)\rightarrow S(E)\) be an isometry. Then \(V_0\) can be extended to a linear isometry defined on the whole \(X\) if and only if the following conditions hold: (i) For every \(x_1,x_2\in S(X)\) and \(\lambda_1,\lambda_2\in \mathbb R\), if \(\| \lambda_1V_0(x_1) + \lambda_2 V_0(x_2)\| =1\), then \(\lambda_1 V_0(x_1) + \lambda_2 V_0(x_2)\in V_0(S(X))\). (ii) For every collection \(\Gamma_1,\Gamma_2,\dots,\Gamma_n\) of mutually disjoint subsets of \(\Gamma\) and every collection \(\lambda_1,\lambda_2,\dots,\lambda_n\) of real scalars, if \(V_0(x)=\sum_{k=1}^n \lambda_k V_0(\chi_{\Gamma_k})\) for some \(x\in S({\mathcal L}^\infty(\Gamma))\), then there exist scalars \(\lambda_1',\lambda_2',\dots,\lambda_n'\) and \(x_0\in X\) with \(\operatorname{supp} x_0 \subset (\bigcup_{k=1}^{n} \Gamma_k)^{\text{c}}\) such that \(x=x_0 + \sum_{k=1}^n \lambda_k' \chi_{\Gamma_k}\). The author also shows that for \(X=c_0\), condition (ii) is redundant and, in particular, if \(V_0:S(c_0)\rightarrow S(E)\) is a surjective isometry, then it can be extended to a linear isometry defined on the whole space \(c_0\). Let us comment that it has recently been shown by \textit{R.~Liu} [J. Math. Anal. Appl. 333, 959--970 (2007; Zbl 1124.46005)] that condition (ii) is also redundant for \(X=\ell_\infty(\Gamma)\) and, in particular, every surjective isometry from \(S(\ell_\infty(\Gamma))\) onto \(S(E)\) can always be extended to be a linear isometry from \(\ell_\infty(\Gamma)\) onto \(E\).
    0 references
    Banach space
    0 references
    isometry
    0 references
    0 references

    Identifiers