Alain Oliviero Durmus is a Professor at École polytechnique and a member of the Centre de Mathématiques Appliquées. His research focuses on computational statistics and machine learning, with a particular interest in Monte Carlo and stochastic methods for Bayesian inference, generative models, and stochastic approximation schemes for optimization and fixed-point problems.

His work spans Monte Carlo methods, Markov chain Monte Carlo methods, variational inference, stochastic optimization, Markov processes, Bayesian statistics, inverse problems, and generative modeling.

Alain Oliviero Durmus is a Hi! PARIS Advanced Fellow for the 2025-2028 period with the project “Towards Enhanced Generative Models.” This fellowship focuses on building a unified mathematical foundation for diffusion models, also known as score-based generative models, through the lens of stochastic optimal control.

The project aims to better understand the theoretical properties of diffusion models, including error bounds, stability, sampling efficiency, and uncertainty quantification, while developing enhanced training and inference strategies with provable guarantees.