Introduction

The full event was filmed an is available here:

About the event

Uncertainty quantification (UQ) employs theoretical, numerical and computational tools to characterise uncertainty. It is increasingly becoming a relevant tool to gain better understanding of physical systems and to make better decisions under uncertainty. Realistic physical systems are usually described by numerical models, often simulated using computer code. The computationally expensive and complex codes can be replaced by inexpensive and functionally simple Gaussian Process (GP) emulators that approximate the functional relationships essential for the purposes of UQ.

This workshop brings together early-career researchers and experts in the field, exploring the theoretical as well as the numerical aspects of GP emulation. Real applications are emphasised, especially those having 'large' features. 'Large' features could include complex physical and numerical models and/or large number of observations or parameters. Specifically, climate, tsunami and earthquake problems are targeted due to their relevance as global challenges.
 

Titles and abstracts 

Abstracts:effective and efficient Gaussian processes 

Speakers

Lassi Roininen

Associate Professor in Applied Mathematics, School of Engineering Science, Lappeenranta-Lahti University of Technology, Finland

Matt Dunlop

Postdoctoral Associate, Courant Institute of Mathematical Sciences

Joakim Beck

PhD, King Abdullah University of science and technology, Saudi Arabia

Finn Lindgren

Chair of Statistics and Bayes Centre Director of Research, The university of Edinburgh

Organisers

Further info

The Alan Turing Institute

First floor British Library, 

96 Euston Road, 

London, NW1 2DB

 

Location

The Alan Turing Institute

1st floor of the British Library, 96 Euston Road, London, NW1 2DB