BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Treed Gaussian Processes for Regression and Classification - Tamar
 a Broderick (University of Cambridge)
DTSTART:20080421T101500Z
DTEND:20080421T111500Z
UID:TALK11274@talks.cam.ac.uk
CONTACT:Carl Scheffler
DESCRIPTION:A Gaussian process is a popular nonparametric model for regres
 sion and classification that specifies a prior over functions. It is often
  constructed so that this distribution over functions is stationary althou
 gh many data sets exhibit only local stationarity. A treed Gaussian proces
 s is an efficient nonstationary modeling scheme that fits stationary Gauss
 ian processes to regions of a treed partition. This partition not only all
 ows a more general\, interpretable model but also facilitates faster inter
 nal Gaussian process calculations\, as for prediction. The treed model has
  recently been developed and applied to regression problems. Here\, it is 
 extended to solve classification problems.
LOCATION:TCM Seminar Room\, Cavendish Laboratory\, Department of Physics
END:VEVENT
END:VCALENDAR