Sparsity Promoting Optimal Transport regularization for superresolution and data-driven models
Organization:
Funded by: | UT/EWI |
PhD: |
|
Supervisor:
Daily supervisor: |
|
Collaboration: | -
|
Description:
In this project, we plan to study the properties of solutions of sparsity-promoting models. In particular, we will focus on static and dynamic models regularized with Optimal Transport energies, with the goal of understanding the superresolution properties of Inverse Problems in the presence of noise. We will then apply the developed theory to analyze the dynamics and improve the training of Machine Learning models.
Output:
A general theory for exact sparse representation recovery in convex optimization (2026) Information and Inference, 15(2) . Carioni, M. & Del Grande, L. https://doi.org/10.1093/imaiai/iaag017 A Dual Certificate Approach to Sparsity in Infinite-Width Shallow Neural Networks (2026) [Working paper › Preprint] . ArXiv.org . Grande, L. D. , Brune, C. & Carioni, M. Nonlocal perimeters and variations: Extremality and decomposability for finite and infinite horizons (2026) Calculus of Variations and Partial Differential Equations, 65(2) . Article 39 . Carioni, M. , Del Grande, L. , Iglesias, J. A. & Schönberger, H. https://doi.org/10.1007/s00526-025-03205-0 Nonlocal perimeters and variations: Extremality and decomposability for finite and infinite horizons (2025) [Working paper › Preprint] . Carioni, M. , Grande, L. D. , Iglesias, J. A. & Schönberger, H. Nonlocal perimeters and variations: Extremality and decomposability for finite and infinite horizons (2025) [Working paper › Preprint] . ArXiv.org . Carioni, M. , Grande, L. D. , Iglesias, J. A. & Schönberger, H. https://doi.org/10.48550/arXiv.2502.05149 Pictures:
