PhD Studentship: EPSRC Industrial Doctoral Landscape Award (IDLA) - Probabilistic Numerics and Inverse Problems University of Cambridge in United Kingdom

This PhD studentship at the University of Cambridge is part of the EPSRC Industrial Doctoral Landscape Award (IDLA) and is offered in collaboration with IBM and the Department of Engineering. The research focuses on Probabilistic Numerics and Inverse Problems, particularly those governed by Partial Differential Equations (PDEs) in Earth and planetary systems. The project aims to advance mathematical and computational modelling by explicitly representing and quantifying uncertainty in numerical computation, a cutting-edge approach in probabilistic numerics.

IBM faces practical challenges in large-scale inverse problems, often using Foundation-Model (FM) surrogates. However, in data-scarce settings where FM training is not feasible, the project will explore synthetic data generation via direct PDE solvers. The PhD will investigate how probabilistic numerical methods can enhance, supplement, or replace existing approaches, enabling more principled uncertainty quantification and improved performance in complex geophysical and planetary models.

Applicants should have (or expect to obtain) a good UK Master's degree (or overseas equivalent) in a relevant science subject such as Engineering, Physics, Computer Science, or Mathematics. Candidates must be self-motivated, able to take ownership of their research, and effectively communicate their findings. Required application documents include a short research statement (maximum 1 page), a curriculum vitae, a publication list, and contact details for two referees. There is a £20 application fee.

EPSRC IDLA studentships are available for eligible home students and a limited number of international students. The University of Cambridge actively supports equality, diversity, and inclusion, and encourages applications from all sections of society. Early applications are encouraged as the position may be filled once suitable candidates are identified. For queries, contact Professor Mark Girolami at [email protected] with a copy to [email protected].

Applications should be submitted via the University Application Portal. The position offers a unique opportunity to work at the intersection of mathematics, computer science, and Earth sciences, contributing to innovative research in uncertainty quantification and computational modelling.