Patrick graduated with an MMath from the University of Warwick in 2021
Patrick graduated with an MMath from the University of Warwick in 2021, with a master’s thesis entitled “Physics Informed Neural Networks (PINNs) for solving Geometric Partial Differential Equations”. His Mathematical interests include deep learning, PDEs, SDEs, their numerical approximations, and Numerical Analysis in general. Outside of Maths, he enjoys football, going to the gym, coding, and playing chess (even though he isn’t very good at it).
Research project title: Learning to optimise in inverse problems
Supervisor(s): Matthias Ehrhardt, Mohammad Golbabaee
Project desription: Inverse Problems focus on recovering information from potentially noisy data. Many such examples exist due to modern imaging modalities, such as MRI and PET. Commonly, inverse problems are tackled by considering variational regularisation. Often, the solution to such minimisation problems is found using iterative methods, such as PDHG and ADMM. This PhD will focus on learning to optimise, which seeks to leverage data to speed up this optimisation process. There are many potential areas for research, from learning hyperparameters in an iterative procedure, to learning how to sample in Stochastic optimisation methods.
Students joining SAMBa in 2022