Badr-Eddine Chérief-Abdellatif

About me

I am a CNRS researcher (Chargé de Recherche) based at Sorbonne Université, in the Laboratoire de Probabilités, Statistique et Modélisation (LPSM). I am also a Maître de Conférence at Ecole Nationale des Ponts et Chaussées.

Before joining CNRS, I was a postdoctoral researcher in the Department of Statistics at the University of Oxford, and I received my PhD in 2020 from Institut Polytechnique de Paris, prepared in the Center for Research in Economics and Statistics (CREST), Paris, under the supervision of Pierre Alquier.

My research focuses on the foundations of machine learning, with a particular interest in learning under uncertainty. I develop principled inference methods that aim to make learning algorithms more reliable and trustworthy, supported by rigorous theoretical guarantees.

Inspired by Bayesian inference, my work develops broader principles for uncertainty quantification and information processing in modern machine learning. I combine ideas from statistics, optimization, and learning theory, with current interests including generalized Bayesian inference, variational inference, PAC-Bayes theory, uncertainty quantification, imprecise probabiliy theory, missing data analysis, kernel methods, and robust machine learning.

You can find here my PhD thesis.

Publications

Preprints

• Asymptotics of Nonparametric Estimation under general non-monotone MAR missingness: A Bayesian Approach.
  B.-E. C.-A., J. Näf
  Preprint ArXiv, 2026.
  

• Maxitive Donsker–Varadhan Formulation for Possibilistic Variational Inference.
  J. Singh, S. Wongso, J. Houssineau, B.-E. C.-A.
  Preprint ArXiv, 2025.
  

• Predictively Oriented Posteriors.
  Y. McLatchie, B.-E. C.-A., D.T. Frazier, J. Knoblauch
  Preprint ArXiv, 2025.
  

• Parametric MMD Estimation with Missing Values: Robustness to Missingness and Data Model Misspecification.
  B.-E. C.-A., J. Näf
  Preprint ArXiv, 2025.
  

• On the Asymptotics of Importance Weighted Variational Inference.
  B.-E. C.-A., R. Douc, A. Doucet, H. Marival
  Preprint ArXiv, 2025.
  

Journal

• Bayes meets Bernstein at the Meta Level: an Analysis of Fast Rates in Meta-Learning with PAC-Bayes.
  C. Riou, P. Alquier, B.-E. C.-A.
  Journal of Machine Learning Research (JMLR), 2025.
  

• Estimation of Copulas via Maximum Mean Discrepancy.
  P. Alquier, B.-E. C.-A., A. Derumigny, J.-D. Fermanian
  Journal of the American Statistical Association (JASA), 2023.
  The paper comes with an R package: MMDCopula .
  

• Finite sample properties of parametric MMD estimation: robustness to misspecification and dependence.
  B.-E. C.-A., P. Alquier
  Bernoulli, 2022.
  

• Consistency of Variational Bayes Inference for Estimation and Model Selection in Mixtures.
  B.-E. C.-A., P. Alquier
  Electronic Journal of Statistics, 2018.

Conference

• From Bayes’ Rule to Bayes Rules: Optimal Information Processing and Axiomatic Foundations Beyond Probability.
  J. Houssineau, B.-E. C.-A.
  Accepted to the 42nd Conference on Uncertainty in Artificial Intelligence (UAI), 2026.
  

• Recursive PAC-Bayes: A Frequentist Approach to Sequential Prior Updates with No Information Loss.
  Y.S. Wu, Y. Zhang, B.-E. C.-A., Y. Seldin
  Accepted to the 38th Conference on Neural Information Processing Systems (NeurIPS), 2024.
  Spotlight.

• Label shift quantification with robustness guarantees via distribution feature matching.
  B. Dussap, G. Blanchard, B.-E. C.-A.
  Accepted to the European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases (ECML PKDD), 2023.
  Best Student Paper Award.

• On PAC-Bayesian reconstruction guarantees for VAEs.
  B.-E. C.-A., Y. Shi, A. Doucet, B. Guedj
  Accepted to the 25th International Conference on Artificial Intelligence and Statistics (AISTATS), 2022.
  

• Convergence Rates of Variational Inference in Sparse Deep Learning.
  B.-E. C.-A.
  Proceedings of the 37th International Conference on Machine Learning (ICML), 2020.
  

• MMD-Bayes: Robust Bayesian Estimation via Maximum Mean Discrepancy.
  B.-E. C.-A., P. Alquier
  Proceedings of the 2nd Symposium on Advances in Approximate Bayesian Inference (AABI), 2020.
  

• A Generalization Bound for Online Variational Inference.
  B.-E. C.-A., P. Alquier, M.E. Khan
  Proceedings of the 11th Asian Conference on Machine Learning (ACML), 2019.
  Best Paper Award ACML 2019.

• Consistency of ELBO maximization for model selection.
  B.-E. C.-A.
  Proceedings of the 1st Symposium on Advances in Approximate Bayesian Inference (AABI), 2019.
  Travel Award.

Patent

• Method for Determining Proportions of Populations in an Ensemble of Biological Objects.
  Inventors: B. Dussap, G. Blanchard, B. Labarthe, B.-E. C.-A.
  Applicants: Metafora Biosystems, Université Paris-Saclay, CNRS
  Published under the Intenational Patent Cooperation Treaty (PCT), 2024.
  The invention relates to the field of cytometric data analysis, including single data analysis.

Teaching

Lecturer 2023-...

• Sorbonne Université -- M2 - PAC-Bayes theory & Variational inference.
• Université Paris Cité -- M2 - High-Dimensional Statistics.
• Ecole des Ponts ParisTech -- 2A - High-Dimensional Statistics.
• ESSEC Business School -- 1A - Business Statistics & Analytics.

Short CV

Short CV

Academic Position

• 2022-........ -- Chargé de Recherche -- CNRS
• 2020-2022 -- Postdoc -- University of Oxford
• 2017-2020 -- PhD -- Institut Polytechnique de Paris