Diameter-preserving maps on various classes of function spaces (Q4548823)

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 1789516
Language Label Description Also known as
default for all languages
No label defined
    English
    Diameter-preserving maps on various classes of function spaces
    scientific article; zbMATH DE number 1789516

      Statements

      Diameter-preserving maps on various classes of function spaces (English)
      0 references
      0 references
      0 references
      27 August 2002
      0 references
      function space
      0 references
      diameter-preserving linear bijection
      0 references
      Let \(X\) be a compact Hausdorff space and for any Banach space \(E\), let \(C(X,E)\) denote the space of \(E\)-valued continuous functions on \(X\), equipped with the supremum norm. A closed point-separating subspace \(A\subset (X,E)\) is called a function space. A map \(T:A\to A\) is said to be diameter preserving if for any \(a\in A\), the sets \(a(X)\) and \(T(a)(X)\) have the same diameter in \(E\). Motivated by a description of such maps obtained in the scalar-valued case by \textit{M. Györy} and \textit{L. Molnar} [Arch. Math. 71, 301-310 (1998; Zbl 0928.46034)] and \textit{F. González} and \textit{V. V. Uspenskij} [Extr. Math. 14, 19-29 (1999; Zbl 0953.54021)], \textit{A. K. Roy} and the reviewer have given, for certain spaces \(E\), a description of such linear, bijective maps when \(A=C(X,E)\) in terms of homeomorphisms of \(X\), isometries of \(E\) and a linear (not necessarily continuous) operator from \(C(X,E)\) to \(E\) [J. Austr. Math. Soc. 70, 323-335 (2001; Zbl 1027.46009)]. In the present paper the authors consider an analogous description of a bijection \(T\) that is not necessarily linear, between spaces \(A_1\subset C(X_1,E_1)\) and \(A_2\subset C(X_2,E_2)\). To compensate for the linearity they impose the conditions \(T(0) =0\), the sets \(a(X_1)- b(X_1)\) and \(T(a)(X_2)- T(b)(X_2)\) have the same diameter for all \(a,b\). In the complex case it is also assumed that \(T(ia)= iT(a)\) for all \(a\). Assuming a linear bijection \(J:E_1\to E_2\), they show that \(T=\overline T+\varphi\) where \(\overline T\) is a diameter-preserving linear bijection and \(\varphi: A_1\to E_2\) is such that \(\varphi(0)=0\). \(a\to a+J^{-1} (\varphi(a))\) is one-to-one and the equation \(J(x)+\varphi(a+x)=0\) is solvable for all \(x\in E_1\) (in the complex case one also has \(\varphi(ia)=i\varphi(a))\). When \(E\) is the scalar field and the function spaces contain constants, it is proved that the existence of a diameter-preserving linear bijection is equivalent to the existence of an onto affine homeomorphism \(\varphi\) between the absolute convex hulls of \(S_{A_2}-S_{A_2}\) and \(S_{A_1}-S_{A_1}\) such that \(\varphi(ix)= i\varphi(x)\). Here \(S_A\) stands for state space. Reviewer's note: In the introduction, the authors use the letters \(Q\) and \(X\) for a compact space and Banach space, respectively; however, in subsequent sections \(X\) and its suffix stand for compact spaces and \(E\) and its suffix for Banach spaces.
      0 references

      Identifiers

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