BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Hidden Common Cause Relations in Relational Learning - Ricardo Sil
 va (Statistical Laboratory)
DTSTART:20071017T130000Z
DTEND:20071017T140000Z
UID:TALK8574@talks.cam.ac.uk
CONTACT:Zoubin Ghahramani
DESCRIPTION:When predicting class labels for objects within a relational d
 atabase\, it is often helpful to consider a model for relationships: this 
 allows for information between class labels to be shared and to improve pr
 ediction performance. However\, there are different ways by which objects 
 can be related within a relational database. One traditional way correspon
 ds to a Markov network structure: each existing relation is represented by
  an undirected edge. This encodes that\, conditioned\non input features\, 
 each object label is independent of other object labels given its neighbor
 s in the graph. However\, there is no reason why Markov networks should be
  the only representation of choice for symmetric dependence structures. He
 re we discuss the case when relationships are postulated to exist due to h
 idden common causes.  We discuss how the resulting graphical model differs
  from Markov networks\, and how it describes\ndifferent types of real-worl
 d relational processes. A Bayesian nonparametric classification model is b
 uilt upon this graphical representation and evaluated with several empiric
 al studies.\n\nJoint work with Wei Chu and Zoubin Ghahramani.\n
LOCATION:LR6\, Engineering\, Department of
END:VEVENT
END:VCALENDAR