Central points and measures and dense subsets of compact metric spaces (Q1947835)
From MaRDI portal
scientific article
Language | Label | Description | Also known as |
---|---|---|---|
English | Central points and measures and dense subsets of compact metric spaces |
scientific article |
Statements
Central points and measures and dense subsets of compact metric spaces (English)
0 references
26 April 2013
0 references
The author works in a category \(\mathcal K\) of metric spaces \(X,Y\) with isometries as morphisms, denoted by Iso\(_{\mathcal K}(X,Y).\) A distinguished point in \(X\in\mathcal K\) is a point \(c_X\in X\) that is preserved by all morphisms, that is, \(\Phi(c_X)=c_Y\) for all \(\Phi\in\) Iso\(_{\mathcal K}(X,Y).\) The author introduces the notion of weak convexity of a metric space, generalizing some known convexity notions in metric spaces (Takahashi, Kijima, Yang and Zhang), and proves that any weakly convex compact metric space has a unique generalized Chebyshev center (defined through an iterative procedure) which is a distinguished point of the space. Since compact convex subsets of normed spaces are weakly convex, this result is related to Kakutani's common fixed point theorem for families of affine mappings. He also studies the existence of the generalized Chebyshev center \(\mu_X\) in the space of probability measures on a compact metric space \(X\) and proves that it is the unique measure invariant under every isometry of \(X\), provided the isometry group of \(X\) acts transitively on \(X\). If \(G\) is a compact metric group, then \(\mu_G\) is the Haar measure on \(G\). The same problem is examined in a class of compact metric spaces, called quasi-nilpotent, and in the last section (Section 6) one studies dense distinguishing sets in compact metric spaces, a notion related to the theory of random metric spaces.
0 references
Chebyshev center
0 references
convex metric space
0 references
Kantorovich metric
0 references
distinguished point
0 references
common fixed point
0 references
Haar measure
0 references
random metric space
0 references