Benjamin Guedj
Benjamin Guedj works on the mathematical foundations of machine learning. His research includes PAC-Bayes theory, computational statistics and shared work on risk measures. Guedj pursued doctoral studies at Sorbonne and Télécom Paris from 2011 to 2013 with Gérard Biau and Eric Moulines. He joined Inria in 2014 and UCL in 2018. He held research-scientist and principal-research-fellow roles before his later titles. With Zakaria Mhammedi and Bob Williamson, he co-authored PAC-Bayesian Bound for the Conditional Value at Risk. The paper received a spotlight presentation at NeurIPS 2020.
- Works
- 120
- Citations
- 516
- h-index
- 11
Publication numbers from OpenAlex, as of 14 October 2026.
Timeline
-
2011–2013
Mathematical-statistics doctorate
Guedj pursued doctoral studies at Sorbonne and Télécom Paris from 2011 to 2013, with Gérard Biau and Eric Moulines.
-
2014–2018
Inria and UCL
He joined Inria in 2014 and UCL in 2018. He held research-scientist and principal-research-fellow roles before his later titles.
-
2020
PAC-Bayes and risk
With Zakaria Mhammedi and Bob Williamson, he co-authored PAC-Bayesian Bound for the Conditional Value at Risk. The paper received a spotlight presentation at NeurIPS 2020.
Sources and credits
Photo: bguedj.github.io; creator not stated. License: Not stated by source. Source: bguedj.github.io. Cropped to a square and resized. Ask us to remove or replace this photo.