A LOOK AT MATTHIEU'S BACKGROUND

Matthieu Muller received the M.Sc. degree in Applied and Industrial Mathematics from the University Grenoble-Alpes, France in 2022, along with an Engineering degree from the Grenoble-INP ENSIMAG, France engineering school the same year.
He obtained in 2025 a joint Ph.D. degree from the Grenoble Institute of Technology (Grenoble-INP),, France at the GIPSA-LAB and the University Of Iceland, Reykjavík, Iceland.
He is currently a post-doc at the Laboratory Hubert Curien, University Jean Monnet, Saint-Étienne, France, working on continuous image representations for inverse problems in image processing.
His research interests include signal and image processing, deep learning, hybrid methods, and self-supervised learning.

MATTHIEU'S MOTIVATION FOR SCIENTIFIC RESEARCH

My research interests lie at the intersection of inverse problems, mathematical optimization, and learning frameworks, making this project a natural fit. I am particularly drawn to approaches that embed principled structure into the learning process, and the challenge of reconstructing high-quality images from degraded observations strikes me as both mathematically rich and practically impactful.
A key appeal of this project is its proposal to represent physical scenes as functions disconnected from sensor grids, enabling finer representations of reality. More precisely, extending unfolded proximal networks to this setting bridges variational methods and operator learning yielding both theoretical depth and strong performance. The resulting discretization-invariance property addresses limitations of current methods, while the self-supervised training strategy ensures applicability to realistic experimental settings where ground-truth data is unavailable.
Beyond the mathematical challenges, the variety of potential applications, from solar to biological imaging, would allow me to develop versatile expertise across scientific domains.