Tom Devynck

I’m a PhD student in deep learning at Université Paris-Saclay, between the DAVID lab (UVSQ) and LISN. I work on methods that help deep networks, from CNNs to vision transformers, focus on the features that matter, to make them more robust and easier to interpret.

I hold a Master’s degree in Econometrics and Statistics from Toulouse School of Economics, and a double bachelor’s degree in Applied Mathematics and Economics from the same school.

On the side, I co-founded Tensaure, a startup born from a Google hackathon, which builds optimisation models for companies.