BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Constrained Optimization and Calibration for Deterministic and Sto
 chastic Simulation Experiments - Lee\, H (University of California\, Santa
  Cruz)
DTSTART:20110908T133000Z
DTEND:20110908T140000Z
UID:TALK32714@talks.cam.ac.uk
CONTACT:Mustapha Amrani
DESCRIPTION:Optimization of the output of computer simulators\, whether de
 terministic or stochastic\, is a challenging problem because of the typica
 l severe multimodality. The problem is further complicated when the optimi
 zation is subject to unknown constraints\, those that depend on the value 
 of the output\, so the function must be evaluated in order to determine if
  the constraint has been violated. Yet\, even an invalid response may stil
 l be informative about the function\, and thus could potentially be useful
  in the optimization. We develop a statistical approach based on Gaussian 
 processes and Bayesian learning to approximate the unknown function and to
  estimate the probability of meeting the constraints\, leading to a sequen
 tial design for optimization and calibration. \n
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR