I am a PhD student in Machine Learning / Statistics at University Paris-Saclay, affiliated with the Institut de Mathématiques d’Orsay and part of the Datashape team at INRIA. My research is conducted under the supervision of Gilles Blanchard and Marc Glisse, and in collaboration with Metafora, a biotechnology company based at the Cochin Hospital.
My thesis focuses on the comparison of cytometric datasets, particularly in the context of Metafora's software, Metaflow. Metaflow enables the automatic analysis of flow cytometry data. My work involves leveraging machine learning models to transfer analysis from one sample to a new, unanalyzed one. This process relies on Reproducing Kernel Hilbert space to embed and store high-dimensional features in Euclidean space. The goal is twofold: estimating the proportions of each population in a new sample and automatically naming the clusters obtained by the software.
- Machine Learning
- Label Shift and Quantification Learning
- Kernel Mean Embedding and kernel methods in general
- "Label Shift Quantification with Robustness Guarantees via Distribution Feature Matching" (with G. Blanchard and B. Chérief-Abdellatif) - ArXiv preprint. This paper was published at ECML/PKDD 2023 and obtained the Research Tracks – Best Student Paper Award. Proceedings.
- Journées de Statistique de la Société Française de Statistique, 2023.
- DataShape Seminar, 2023.
- Séminaire des doctorants de l'équipe Probabilité et Statistiques de l'Institut de Mathématiques d'Orsay, 2023.
- ECML/PKDD 2023, Label Shift Quantification with Robustness Guarantees via Distribution Feature Matching (RT Track – Best Student Paper).
- Workshop Efficient Statistical Testing for high-dimensional model (FAST-BIG).
- ECML/PKDD 2023, Label Shift Quantification with Robustness Guarantees via Distribution Feature Matching (RT Track – Best Student Paper).
I co-organize a seminar for master students in Statistics and Machine Learning at Université Paris-Saclay.