BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//132.216.98.100//NONSGML kigkonsult.se iCalcreator 2.20.4//
BEGIN:VEVENT
UID:20250712T114823EDT-921399ro8z@132.216.98.100
DTSTAMP:20250712T154823Z
DESCRIPTION:\n	Abstract:\n\n\nA recent body of work\, by myself and many oth
 ers\, aims to develop a statistical theory of network data for problems a 
 single network is observed. Of the models studied in this area\, graphon m
 odels are probably most widely known in statistics. I will explain the rel
 ationship between three aspects of this work: (1) Specific models\, such a
 s graphon models\, graphex models\, and edge-exchangeable graphs. (2) Samp
 ling theory for networks\, specifically in the case statisticians might re
 fer to as an infinite-population limit. (3) Invariance properties\, especi
 ally various forms of exchangeability. I will also present recent results 
 that show how statistically relevant results (such as central limit theore
 ms) can be derived from such invariance properties.\n\n\n	Speaker\n\n\nPete
 r Orbanz is an Associate Professor in the Department of Statistics at Colu
 mbia University. He studies network and relational data\, Bayesian nonpara
 metrics\, asymptotics and ergodicity of random structures\, and hierarchie
 s of latent variables. More generally\, he is interested in all mathematic
 al aspects of machine learning and artifical intelligence.\n\nHe was an un
 dergraduate at the University of Bonn\, a PhD student of Joachim M. Buhman
 n at ETH Zurich\, and a postdoc of Zoubin Ghahramani at the University of 
 Cambridge.\n
DTSTART:20190201T203000Z
DTEND:20190201T213000Z
LOCATION:Room 1205\, Burnside Hall\, CA\, QC\, Montreal\, H3A 0B9\, 805 rue
  Sherbrooke Ouest
SUMMARY:Peter Orbanz (Columbia University)
URL:/mathstat/channels/event/peter-orbanz-columbia-uni
 versity-293826
END:VEVENT
END:VCALENDAR