BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Bayesian optimal design for Gaussian process model - Maria Adamou 
 (University of Southampton)
DTSTART:20180208T160000Z
DTEND:20180208T170000Z
UID:TALK100225@talks.cam.ac.uk
CONTACT:INI IT
DESCRIPTION:<span>Co-author: Dave Woods		(University of Southampton)      
   <br></span><br>Data collected from correlated processes arise in many di
 verse application areas including both computer and physical experiments\,
  and studies in environmental science. Often\, such data are used for pred
 iction and optimisation of the process under study. For example\, we may w
 ish to construct an emulator of a computationally expensive computer model
 \, or simulator\, and then use this emulator to find settings of the contr
 ollable variables that maximise the predicted response.  The design of the
  experiment from which the data are collected may strongly influence the q
 uality of the model fit and hence the precision and accuracy of subsequent
  predictions and decisions. We consider Gaussian process models that are t
 ypically defined by a correlation structure that may depend upon unknown p
 arameters. This parametric uncertainty may affect the choice of design poi
 nts\, and ideally should be taken into account when choosing a design. We 
 consider a decision-theoretic Bayesian design for Gaussian process models 
 which is usually computationally challenging as it requires the optimizati
 on of an analytically intractable expected loss function over high-dimensi
 onal design space. We use a new approximation to the expected loss to find
  decision-theoretic optimal designs. The resulting designs are illustrated
  through a number of simple examples.
LOCATION:Seminar Room 1\, Newton Institute
END:VEVENT
END:VCALENDAR