There are potentially countless applications of deep learning to a vast number of problems. We cannot hope to look at all of them in this project, but we will study some areas of significant importance using the theory and algorithms developed in the other two areas of the programme.
Applications will focus on efficient modelling and inversion of large complex systems in space and time, including medical imaging, weather forecasting, and other environmental problems. These applications will be informed by, and will test the underlying theory and algorithms above. Work Packages in this area will closely involve many of our External Partners including BT, GE Healthcare, and the Met Office.