BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Batch sequential experimental designs for computer experiments - N
 otz\, W (Ohio State University)
DTSTART:20110906T143000Z
DTEND:20110906T150000Z
UID:TALK32669@talks.cam.ac.uk
CONTACT:Mustapha Amrani
DESCRIPTION:Finding optimal designs for computer experiments that are mode
 led using a stationary Gaussian Stochastic Process (GaSP) model is challen
 ging because optimality criteria are usually functions of the unknown mode
 l parameters. One popular approach is to adopt sequential strategies. Thes
 e have been shown to be very effective when the optimality criterion is fo
 rmulated as an expected improvement function. Most of these sequential str
 ategies assume observations are taken sequentially one at a time. However\
 , when observations can be taken k at a time\, it is not obvious how to im
 plement sequential designs. We discuss the problems that can arise when im
 plementing batch sequential designs and present several strategies for seq
 uential designs taking observations in k-at-a-time batches. We illustrate 
 these strategies with examples.\n
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR
