BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Enhancing Stochastic Kriging Metamodels with Stochastic Gradient E
 stimators - Staum\, J (Northwestern University)
DTSTART:20110907T103000Z
DTEND:20110907T110000Z
UID:TALK32691@talks.cam.ac.uk
CONTACT:Mustapha Amrani
DESCRIPTION:Stochastic kriging is the natural extension of kriging metamod
 els for the design and analysis of computer experiments to the design and 
 analysis of stochastic simulation experiments where response variance may 
 differ substantially across the design space. In addition to estimating th
 e mean response\, it is sometimes possible to obtain an unbiased or consis
 tent estimator of the response-surface gradient from the same simulation r
 uns. However\, like the response itself\, the gradient estimator is noisy.
  In this talk we present methodology for incorporating gradient estimators
  into response surface prediction via stochastic kriging\, evaluate its ef
 fectiveness in improving prediction\, and specifically consider two gradie
 nt estimators: the score function/likelihood ratio method and infinitesima
 l perturbation analysis. \n
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR