Project overview
This project tackles problems in uncertainty quantification (UQ). Statistical UQ uses design, modelling and inference methods to describe, understand, classify and, where appropriate, minimise uncertainty when applying complex mathematical/computational models. These aims are achieved by answering questions such as: at which input combinations should we run expensive computer models? What (often limited) physical data should be collected? How can we combine these (simulation, physical) data sources to learn about (physically meaningful) model parameters, and to make calibrated (bias-adjusted) predictions of the system under study?
Staff
Lead researcher
