Convex functions and \(p\)-barycenter on CAT(1)-spaces of small radii (Q1674145): Difference between revisions

A \(p\)-barycenter of a probability measure \(\mu\) on a space \(X\) is a global minimum of the function \(x\to\int_Xd^p(.,x)d\mu\). The paper under review proves for \(\mathrm{CAT}(\kappa)\)-spaces \(X\) with \(\kappa>0\) and for every \(p\geq 2\) the existence of a \(p\)-barycenter for any probability measure \(\mu\) concentrated on a ball of radius \(\frac{\pi}{2\sqrt{\kappa}}\). Among the consequences are results of Banach-Saks-Kakutani type, saying for example that for a sequence of points in a complete \(\mathrm{CAT}(\kappa)\)-space the \(p\)-barycenters of some subsequence converge, and also a new proof of \(p\)-uniform convexity for \(\mathrm{CAT}(0)\)-spaces.