Hind ATBIR

About Me

Since October 2024, I am a PhD student in statistical machine learning at Jean Monnet University. I work at Hubert Curien Laboratory in the Data Intelligence Team. My thesis, entitled Learning Fair and Robust Kernel-Based Models with Generalization Guarantees, is funded by a PhD scholarship from the French Ministry of Higher Education and Research and supported by the ANR project FAMOUS. I am supervised by Farah Cherfaoui, Rémi Eyraud, Emilie Morvant, and Paul Viallard. Our work aims to provide theoretical guarantees (PAC-Bayesian generalization bounds) on the performance of machine learning models that account for fairness and/or robustness.

Publications

• Une borne PAC-Bayésienne sur une mesure de risque pour l'apprentissage équitable
  Hind Atbir, Farah Cherfaoui, Guillaume Metzler, Emilie Morvant, Paul Viallard
  Conférence sur l’Apprentissage automatique (CAp), 2024

Seminars

• March 2025 : ANR project Famous workshop, Université d’Aix-Marseille, France
  A PAC-Bayesian Bound on a Measure of Risk
• October 2024 : Seminar of the MALICE Team, Inria F-42023, Saint Etienne, France
  PAC-Bayesian Generalization Bound for Fairness

Teaching

• 2025
• Master 1 Informatique – Machine Learning Fundamentals : Lecture Decision Tree, Lab Instrutor
• Master 1 Erasmus Mundus COSI/IMLEX – Machine Learning Fundamentals and Algorithms : Lab Instrutor
• Licence 1 Informatique – Programmation Impérative : Lab Instrutor