Valentin De Bortoli
Research Scientist at Google DeepMind
Paris, Ile-de-France
Summary
👤
Senior🎓
Top SchoolValentin De Bortoli is a research scientist specializing in stochastic methods for generative modelling, currently at Google DeepMind and on leave from CNRS (ENS Ulm, Paris). With nine years of experience spanning a PhD and postdoc at leading institutions including Oxford, he blends rigorous theoretical work with practical deep learning research for data generation. His background in high-level mathematics (Agrégation, top rankings) and a strong MVA training informs precise probabilistic formulations and sampling techniques. Valentin’s work bridges statistics and machine learning, focusing on both provable properties and scalable algorithms—an approach reflected in publications tracked on his Google Scholar and personal site. He often explores non-local image statistics and sampling ideas that transfer from theory to modern generative models.
9 years of coding experience
Classe préparatoire aux grandes écoles aux LazaristesLicentiate degree, Mathématiques, Licentiate degree, Mathématiques at ENS CachanMaster's degree, Mathématiques, Master's degree, Mathématiques at Université Paris DiderotMVA Master Vision Apprentissage, High honors, MVA Master Vision Apprentissage, High honors at École Normale Supérieure Paris-Saclay
Github Skills (10)
generative8
manifold7
stochastic-processes7
image-viewer6
viewer6
image5
image-processing4
jekyll-theme4
svg3
jekyll3
Programming languages (4)
Github contributions (5)
diffusionworkshop/diffusionworkshop.github.io
Aug 2023 - Aug 2023
Contributions:3 PRs, 54 pushes, 10 branches in 28 days
vdeborto/custom-tree
Sep 2022 - Jan 2023
Contributions:4 pushes, 1 branch in 3 months
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.