Juliette Marrie

I am a 3rd-year PhD student at NAVER LABS Europe and INRIA Grenoble - Rhône-Alpes, THOTH team, under the supervision of Julien Mairal, Diane Larlus, and Michael Arbel. I graduated from Mines Paris with a focus in Applied Mathematics and obtained a Master’s Degree in Mathematics, Machine Learning, and Computer Vision (MVA) from ENS Paris-Saclay (2021).

My current research focuses on visual representation learning, with an emphasis on leveraging large pretrained visual models for applications constrained by computational resources or data scarcity. I have worked on automatically learning data augmentation for supervised tasks, and on transferring generic representations from large pretrained models into smaller ones (knowledge distillation) and into 3D Gaussian Splatting scenes.

Professional experience

• 2021: 6-month internship at Philips Research (France) on self-supervised learning from 3D medical images.

• 2020: 6-month internship at Weill Cornell Medicine and the NYGC (Landau Lab, New York) on phylogenetic tree reconstruction from microsatellite sequences.

• 2019: 6-month internship at Neural Concept (start-up in EPFL) on 3D shape optimization using Geometric Deep Learning

Teaching

African Masters of Machine Intelligence (AMMI) 2023 and 2024 - Saly, Senegal
I served as a teaching assistant for two-week courses on Kernel Methods, taught by Jean-Philippe Vert, in July 2023 and July 2024.