About me
I am an associate professor (maître de conférences HDR) at Université Paris-Saclay, in the A&O team (Learning and Optimization) of the LISN, joint with the Inria TAU team. I work at the interface of signal processing, inverse problems and machine learning.
My research started with sparse and structured approximations (mixed norms, iterative thresholding, time-frequency representations). I now focus on learning for inverse problems: unrolled algorithms, dictionary learning, plug-and-play methods and learning without ground truth. The underlying question remains the same: what a (learned) prior really identifies, and what the measurements leave undetermined. Applications range from audio source separation to functional neuroimaging (M/EEG, epilepsy).
Before joining LISN in 2022, I spent 13 years in the Inverse Problems Group of L2S. I received the engineering degree from UTC and the MVA master from ENS Cachan (2005), a PhD in applied mathematics from Université de Provence (2008), and the HDR from Université Paris-Sud (2018). I also head the third year (L3) of the Engineering Sciences (SPI) double degree.
