Eric Moulines

Éric Moulines is a researcher in statistical learning, numerical probability and statistical signal processing. He has devoted a large part of his career to Bayesian inference, Monte Carlo methods and the analysis of stochastic algorithms. His recent work focuses on the challenges of high-dimensional statistical inference, Monte Carlo sampling, inverse problems with diffusive a priori and stochastic optimization. In particular, he explores the links between diffusion methods and Monte Carlo, develops non-asymptotic bounds for stochastic approximation, and designs probabilistic frameworks for generative models and predictive uncertainty. His contributions also cover collaborative and federated learning, and applications of probabilistic models to deep learning and high-dimensional inverse problems. He is at the interface between probability theory and modern statistical learning.

--> International journals with reading commitees references

--> Articles in the proceedings of international conferences references

Page last updated: December 15, 2025