BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Co-clustering of non-smooth graphons - David Choi (Carnegie Mellon
  University)
DTSTART:20160715T143000Z
DTEND:20160715T150000Z
UID:TALK66778@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:Theoretical results are becoming known for community detection
  and clustering  of networks\; however\, these results assume an idealized
  generative model that is  unlikely to hold in many settings. Here we cons
 ider exploratory co-clustering of  a bipartite network\, where the rows an
 d columns of the adjacency matrix are  assumed to be samples from an arbit
 rary population. This is equivalent to  assuming that the data is generate
 d from a nonparametric model known as a  graphon. We show that co-clusters
  found by any method can be extended to the row  and column populations\, 
 or equivalently that the estimated blockmodel  approximates a blocked vers
 ion of the generative graphon\, with generalization  error bounded by n^{-
 1/2}. Analogous results are also shown for degree-corrected  co-blockmodel
 s and random dot product bipartite graphs\, with error rates  depending on
  the dimensionality of the latent variable space.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR