Rémi Flamary is a Professor in the Department of Applied Mathematics and at the Centre de Mathématiques Appliquées at École polytechnique. His research focuses on statistical machine learning and statistical signal processing, with a particular emphasis on optimal transport for machine learning and graph-based methods.

His work spans optimal transport, domain adaptation, transfer and multi-task learning, representation learning, deep learning, variable selection, and optimization. He also works on applications of statistical learning to biomedical signals, brain-computer interfaces, remote sensing, hyperspectral imaging, energy and climate, and astrophysical image processing.

Rémi Flamary is also a Hi! PARIS Synergy Fellow for the 2025-2028 period, together with Florence D’Alché-Buc, Charlotte Laclau, and Karim Lounici, on the project “Advancing Efficient, Reliable and Science-Informed Learning for Non-Euclidean Data with Application to Molecule and Biological Network Structures.” This fellowship focuses on non-Euclidean data, graphs, optimal transport, and representation learning.

His recent work includes research on optimal transport plan prediction between unbalanced graphs, graph-level autoencoders, domain adaptation benchmarks, graph prediction with optimal transport losses, and optimal transport methods for biosignals and graph neural networks.