Research

Research interest

My field of research is in Applied Mathematics, more specifically within the field of Scientific Machine Learning and Uncertainty Quantification. My foundation is linked to the PDE community and I am interested in connecting data-driven learning models with PDE-numerical solvers, all while taking into account and quantifying the uncertainties, for the purpose of creating robust, explicable and generalizable models. This has numerous applications in engineering tasks, such as the development of numerical twins, industrial component health management and prognostics, or more for scientific applications (e.g climate studies or astronomy). My approach is both theoretical and computational and I strongly believe in the synergy of the two.

Talks

09/2024 I3P Seminar, EDF Lab Chatou - slides (in French)
02/2024 Contributed Paper Session, SIAM-UQ-2024 - slides
12/2023 PhD day, EDF Lab Chatou
10/2023 ETICS 2023, Lège Cap-Ferret
09/2023 CJC-MA 2023, CentraleSupélec - poster
04/2023 MASCOT-NUM 2023, Le Croisic
11/2022 Modelling research group seminar, Centre Borelli, ENS Paris-Saclay
10/2022 PSPP seminar, EDF Lab Chatou

Publications

Jaber, E. and Blot, V. and al., Conformalizing Gaussian Processes For More Robust Uncertainty Quantification, ArXiV 2401.07733, Submitted in Journal of Machine Learing for Modeling and Computing, 2024

Jaber, E. and al., Sensitivity Analyses of a Multi-Physics Long-Term Clogging Model For Steam Generators, ArXiV 2401.05741, Published in International Journal of Uncertainty Quantification, 2024

Misc

Msc. dissertation (in French): Removable singularities for bounded vector field solutions of certain first order PDEs.