Data-driven regularization

Data-Driven Convex Regularizers for Inverse Problems

We learn the regularizer from data and make sure that it is convex, which gives guarrantees on convergence of variational optimisation.

Subhadip Mukherjee, Sören Dittmer, Zakhar Shumaylov, Sebastian Lunz, Ozan Öktem, Carola-Bibiane Schönlieb