Autor(en): Voit, Michael
Titel: Continuous Association Schemes and Hypergroups
Sprache (ISO): en
Zusammenfassung: Classical finite association schemes lead to finite-dimensional algebras which are generated by finitely many stochastic matrices. Moreover, there exist associated finite hypergroups. The notion of classical discrete association schemes can be easily extended to the possibly infinite case. Moreover, this notion can be relaxed slightly by using suitably deformed families of stochastic matrices by skipping the integrality conditions. This leads to a larger class of examples which are again associated to discrete hypergroups. In this paper we propose a topological generalization of association schemes by using a locally compact basis space $X$ and a family of Markov-kernels on $X$ indexed by some locally compact space $D$ where the supports of the associated probability measures satisfy some partition property. These objects, called continuous association schemes, will be related to hypergroup structures on $D$. We study some basic results for this notion and present several classes of examples. It turns out that for a given commutative hypergroup the existence of an associated continuous association scheme implies that the hypergroup has many features of a double coset hypergroup. We in particular show that commutative hypergroups, which are associated with commutative continuous association schemes, carry dual positive product formulas for the characters. On the other hand, we prove some rigidity results in particular in the compact case which say that for given spaces $X,D$ there are only a few continuous association schemes.
Schlagwörter: association schemes
Gelfand pairs
hypergroups
spherical functions
positive definite functions
positive product formulas
rigidity results
random walks on association schemes
URI: http://hdl.handle.net/2003/38162
http://dx.doi.org/10.17877/DE290R-20141
Erscheinungsdatum: 2018-02
Enthalten in den Sammlungen:Preprints der Fakultät für Mathematik

Dateien zu dieser Ressource:
Datei Beschreibung GrößeFormat 
Preprint 2018-02.pdfDNB655.3 kBAdobe PDFÖffnen/Anzeigen


Diese Ressource ist urheberrechtlich geschützt.



Alle Ressourcen in diesem Repository sind urheberrechtlich geschützt.