BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Subsampling\, symmetry and averaging in networks - Peter Orbanz (C
 olumbia University)
DTSTART:20160711T133000Z
DTEND:20160711T140000Z
UID:TALK66701@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:Consider a very large graph---say\, the link graph of a large 
 social network. Now invent a randomized algorithm that extracts a smaller 
 subgraph. If we use the subgraph as sample data and perform statistical an
 alysis on this sample\, what can we learn about the underlying network? Cl
 early\, that should depend on the subsampling algorithm. I show how the ch
 oice of algorithm defines a notion of (1) distributional invariance and (2
 ) of averaging within a single large graph. Under suitable conditions\, th
 e resulting averages satisfy a law of large numbers\, such that statistica
 l inference from a single sample graph is indeed possible. From this algor
 ithmic point of view\, graphon models arise from a specific choice of samp
 ling algorithm\, various known pathologies of these models are explained a
 s a selection bias. <br><br>
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR