University of Cambridge > > British Antarctic Survey > Uncertainty quantification for complex simulators using emulation

Uncertainty quantification for complex simulators using emulation

Add to your list(s) Download to your calendar using vCal

If you have a question about this talk, please contact Max Holloway.

Many climate models are complex and expensive, making it difficult to use them in any statistical analysis. For example, standard methods for estimating unknown parameters, doing sensitivity analysis, or quantifying prediction uncertainty, all require the simulator to be run tens of thousands of times. Emulation (also known as surrogate modelling) is the approach of building a cheaper statistical version of your simulator, which approximates its behaviour, but at a fraction of the cost.

In this talk I will introduce Gaussian process emulators, and discuss some recent extensions, particularly in examples with high dimensional inputs and outputs. I will illustrate the approach by discussing recent work on emulating subsurface flow models, as well as highlighting recent work by others in using emulators to train ice sheet models, and to calibrate the NEMO ocean model.

This talk is part of the British Antarctic Survey series.

Tell a friend about this talk:

This talk is included in these lists:

Note that ex-directory lists are not shown.


© 2006-2023, University of Cambridge. Contact Us | Help and Documentation | Privacy and Publicity