Baptiste Ferrere

About Me

I am a Ph.D. student in Applied Mathematics at the Université de Toulouse, in collaboration with EDF R&D, where I am working on explainability and uncertainty quantification for machine learning models in high-stakes applications.

My research is supervised by Fabrice Gamboa and Jean-Michel Loubes (Université de Toulouse), and Nicolas Bousquet and Joseph Muré (EDF R&D).

Before starting my Ph.D., I completed a 6-month research internship at EDF R&D focusing on the generalization of the Hoeffding decomposition for correlated inputs, applied to uncertainty quantification.

I also hold an M.Sc. in Statistical Learning from the Institut Polytechnique de Paris, and an engineering degree from ENSAE Paris, where I specialized in statistics, probability and machine learning.

Research Interests

My research lies at the intersection of applied mathematics and machine learning. I focus on:

• Uncertainty quantification in ML models
• Sensitivity analysis and functional decompositions (e.g. ANOVA, Hoeffding)
• Explainability and interpretability of black-box models
• Theoretical analysis of complex models (e.g., generalization, structure of variance)

I am particularly interested in mathematically grounded methods that improve our understanding of ML models and help build trust in AI systems.