Abstract. We provide a general criterion to deduce maximal amenability of von Neumann subalgebras LΛ ⊂ LΓ arising from amenable subgroups Λ of discrete countable groups Γ. The criterion is expressed in terms of Λ-invariant measures on some compact Γ-space. The strategy of proof is different from S. Popa's approach to maximal amenability via central sequences [Po83], and relies on elementary computations in a crossed-product C * -algebra.
Nous nous intéressons aux groupoïdes discrets préservant une mesure de probabilité, aux actions de groupes et aux relations d'équivalence dans le contexte d'espaces de probabilité généraux. Pour ces objets, nous considérons les notions de coût, de nombres de Betti 2 , d'invariant β et des variantes de dimension supérieure. Nous proposons aussi, sous de faibles hypothèses de finitude, divers résultats de convergence de nombres de Betti 2 et de gradient de rang pour des suites d'actions, de groupoïdes et de relations d'équivalence. En particulier, nous établissons le lien entre le coût combinatoire et le coût de la relation d'équivalence ultralimite. Enfin, nous étudions une version relative de la propriété de Stuck-Zimmer.
We associate to every action of a Polish group on a standard probability space a Polish group that we call the orbit full group. For discrete groups, we recover the wellknown full groups of pmp equivalence relations equipped with the uniform topology. However, there are many new examples, such as orbit full groups associated to measure preserving actions of locally compact groups. In fact, we show that such full groups are complete invariants of orbit equivalence.We give various characterizations of the existence of a dense conjugacy class for orbit full groups, and we show that the ergodic ones actually have a unique Polish group topology. Furthermore, we characterize ergodic full groups of countable pmp equivalence relations as those admitting non-trivial continuous character representations.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.