BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Bayesian Quadrature for Prediction and Optimisation - Michael Osbo
 rne ( Oxford University)
DTSTART:20111128T110000Z
DTEND:20111128T120000Z
UID:TALK34828@talks.cam.ac.uk
CONTACT:Zoubin Ghahramani
DESCRIPTION:Bayesian inference often requires the evaluation of nonanalyti
 c integrals. These must be approximated using methods for numerical integr
 ation\, or quadrature\, of which Monte Carlo techniques are the most commo
 n. A more principled alternative is found in Bayesian Quadrature\, also kn
 own as Bayesian Monte Carlo\, which employs a Gaussian process model for t
 he integrand. We describe an extension of previous Bayesian quadrature tec
 hniques that explicitly models the non-negativity of integrands. We then p
 resent results of the use of Bayesian quadrature for problems related to c
 hangepoint and fault detection\, global optimisation and sensor selection.
LOCATION:Engineering Department\, CBL Room 438
END:VEVENT
END:VCALENDAR