Extremal structure and duality of Lipschitz free spaces (Q1635770)

From MaRDI portal
Revision as of 05:51, 1 March 2024 by RedirectionBot (talk | contribs) (‎Removed claim: author (P16): Item:Q2020392)
scientific article
Language Label Description Also known as
English
Extremal structure and duality of Lipschitz free spaces
scientific article

    Statements

    Extremal structure and duality of Lipschitz free spaces (English)
    0 references
    0 references
    0 references
    0 references
    1 June 2018
    0 references
    Let \(M\) be a pointed metric space with a distinguished point \(0\). Denote by \(\text{Lip}_0(M)\) the space of all Lipschitz functions on \(M\) which vanish at \(0\in M\). Also, denote by \(\delta\) the canonical isometric embedding of \(M\) into \(\text{Lip}_0(M)^*\). The Lipschitz free space \(\mathcal{F}(M)\) (also known as Arens-Eells space) is defined as \(\mathcal{F}(M)=\overline{\text{span}}\{\delta(x): x\in M\}\). A molecule is an element of \(\mathcal{F}(M)\) of the form \[ m_{xy}:=\frac{\delta(x)-\delta(y)}{d(x,y)}, \] where \(x,y\in M\), \(x\neq y\). In the paper under review, the authors study the extremal structure of the closed unit ball \(B_{\mathcal{F}(M)}\). In the 1990s, it was proven by \textit{N. Weaver} [Lipschitz algebras. Singapore: World Scientific (1999; Zbl 0936.46002), Corollary 2.5.4] that every preserved extreme point of \(B_{\mathcal{F}(M)}\) must be a molecule. The main open question in this direction is that whether every extreme point of \(B_{\mathcal{F}(M)}\) is necessarily a molecule, too. For the following chain of implications \[ \text{strongly exposed}\overset{(1)}{\Rightarrow}\text{denting}\overset{(2)}{\Rightarrow}\text{preserved extreme}\overset{(3)}{\Rightarrow}\text{extreme}, \] the authors prove that the converse of \((2)\) holds in general for \(\mathcal{F}(M)\), but the converses of \((1)\) and \((3)\) are both false. However, they also show that the converses of \((1)\) and \((3)\) are true for some special classes of metric spaces. For instance, the converse of \((1)\) holds when \(M\) is a uniformly discrete space and the converse of \((3)\) holds for certain compact spaces \(M\). The authors also introduce and study a distinguished class of preduals of \(\mathcal{F}(M)\), which they call natural preduals. In particular, they prove that, if \(\mathcal{F}(M)\) has such a natural predual and it satisfies some additional assumptions, then the set of extreme points conicides with the set of strongly exposed points. We end this review by remarking that in a recent paper, \textit{R. J. Aliaga} and \textit{A. J. Guirao} [Stud. Math. 245, No. 1, 1--14 (2019; Zbl 1420.46014)] have pushed the study of extremal structure of \(B_{\mathcal{F}(M)}\) even further.
    0 references
    extreme point
    0 references
    dentability
    0 references
    Lipschitz free
    0 references
    duality
    0 references
    uniformly discrete
    0 references

    Identifiers