Workshop on Recent Advances in
Learned Regularisation
20-21 February 2025
66-72 Gower Street, UCL
66-72 Gower Street, UCL
| Home | Programme | Directions |
| 11.00-11.10 | Introductory Remarks |
| 11.10-11.50 | Learning variational regularisations with optimal error estimates |
| Martin Benning | |
| University College London |
| 11.50-12.30 | A bilevel framework for variational image reconstruction with learned convex regularisers |
| Hok Shing Wong | |
| University of Bath |
| 12.30-13.30 | Lunch |
| 13.30-14.10 | Conformal-prediction-based error quantification for image reconstruction with learned priors |
| Martin Holler | |
| University of Graz |
| 14.10-14.50 | Plug-and-play flow matching for regularization |
| Paul Hagemann | |
| Technische Universität Berlin |
| 14.50-15.30 | UNSURE: Unknown noise level Stein's unbiased risk estimator |
| Julian Tachella | |
| Ecole Normale Supérieure de Lyon |
| 15.30-16.00 | Coffee |
| 16.00-16.40 | Bayesian model comparison with learned data-driven priors |
| Jason McEwen | |
| University College London |
| 9.00-9.40 | Iterative refinement of data-adaptive regularization |
| Sebastian Neumayer | |
| Technische Universität Chemnitz |
| 9.40-10.20 | Practical operator sketching framework for accelerating iterative data-driven solutions in inverse problems |
| Billy Junqi Tang | |
| University of Birmingham |
| 10.20-10.40 | Coffee & Tea Break |
| 10.40-11.20 | Proximal operator learning meets unrolling for limited angle tomography |
| Tatiana Bubba | |
| University of Ferrara |
| 11.20-12.00 | Tackling fundamental challenges in hypothesis testing in imaging inverse problems |
| Marcelo Pereyra | |
| Heriot-Watt University |
| 12.00-13.00 | Lunch |
| 13.00-13.40 | Embedding Blake-Zisserman regularization in unfolded proximal neural networks for enhanced edge detection |
| Nelly Pustelnik | |
| Ecole Normale Supérieure de Lyon |
| 13.40-14.20 | Plug-and-play half-quadratic splitting for ptychography |
| Johannes Hertrich | |
| Université Paris Dauphine-PSL |