Bayesian design of experiments for Gaussian process regression
Dave Woods (University of Southampton)
Thursday 14th May, 2015 15:00-16:00 Maths 204
Data collected from correlated processes arise in many diverse application areas including both computer and physical experiments, and studies in environmental and ecological science. Often, such data are used for prediction and optimisation of the process under study. For example, we may wish to construct an emulator of a computationally expensive computer model, or simulator, and then use this emulator to find settings of the controllable variables that maximise the predicted response.
The design of the experiment from which the data are collected may strongly influence the quality of the model fit and hence the precision and accuracy of subsequent predictions and decisions.
We consider Bayesian design for Gaussian process models, and use a new approximation to the expected loss to tackle two problems: (i) decision-theoretic optimal design for prediction; and (ii) robust optimisation of computationally expensive simulators.
The resulting designs are illustrated through a number of simple examples typical of computer experiments and spatial statistics.