Nicolas Courty

Nicolas Courty

Full Professor of Computer Science

Université Bretagne Sud


“One should always be a little improbable.”

Oscar Wilde

Nicolas Courty is Full Professor at University Bretagne Sud since 2018. He obtained his PhD degree in 2002 from INSA Rennes and his 'Habilitation à diriger des recherches' in. 2013, on the topic of computer graphics and animation (avatars, crowds), with a specialization in data-driven methods. He now leads the Obelix team in IRISA, dedicated to machine learning and its applications to Earth Observation. He is an experienced researcher in the domain of machine learning and AI. Among others, he has published several papers in top tier machine learning conferences (NeurIPS, ICLR, ICML, AISTATS, etc.), computer vision (IEEE TPAMI, ECCV, ACCV) and remote sensing (IEEE TGRS, ISPRS journal). From 2014, he has developed an expertise in the domain of optimal transport and related applications to machine learning. He is also one of the recipients of the U.V. Helava Award, awarded by the International Society for Photogrammetry and Remote Sensing (ISPRS), for the best paper in the ISPRS journal in years 2012--2015. Nicolas Courty is a member of the European ELLIS society. From 2020, he pilots an ANR Chair program on AI (OTTOPIA), on the topic of applied optimal transport for Remote Sensing.

Keywords: Optimal transport, Statistical learning, kernel methods, manifold and geometric approaches, optimal control

Recurrent collaborators


  • POT The python Optimal Transport Toolbox

On-going PhD Students and post-doc


I mostly teach computer and data science at Université Bretagne Sud. Among others, I give the following courses:

  • Deep Learning M2 AIDN
  • High performance computing and GPU programming M1 AIDN

I also teach in the Master Erasmus Mundus Copernicus CDE in the GeoData Science track.


We offer some positions at Obelix for master internships, PhD positions or post-docs. Consult the corresponding webpage. Also, if you are interested in doing research with me, and if you have an excellent track record but do not find a suitable announce, feel free me to contact me or use contact widget !

Recent Posts and news

2 new publications at ICML 2021

Two of our papers have made it to ICML 21 ! Unbalanced minibatch Optimal Transport; applications to Domain Adaptation Arxiv Link Online Graph Dictionary Learning" Arxiv Link

POT has made it to JMLR

POT (The Python Optimal Transport toolbox) is a fantastic library for coding optimal transport problems in Python. The associated technical publication has been published in the Software Track of JMLR. Please check [ the POT website].

New publication at Neurips 2020 : Co-Optimal Transport

Optimal transport (OT) is a powerful geometric and probabilistic tool for finding correspondences and measuring similarity between two distributions. Yet, its original formulation relies on the existence of a cost function between the samples of the two distributions, which makes it impractical for comparing data distributions supported on different topological spaces.

One paper accepted at ACCV 2020

One paper accepted at ACCV 2020 on Contextual Semantic Interpretability, with Diego Marcos, Ruth Fong, Sylvain Lobry, Rémi Flamary and Devis Tuia. Paper abstract: Convolutional neural networks (CNN) are known to learn an image representation that captures concepts relevant to the task, but do so in an implicit way that hampers model interpretability.

Our paper on Fused Gromov-Wasserstein published in Algorithms

Our in-depth paper on the Fused-Gromov Wasserstein is published in a special issue of MDPI Algorithms on graphs. And it made the cover ! Check the link to the special issue

On-going Projects

ANR Chair in AI OTTOPIA (2021–2025)

OTTOPIA is a project dedicated to novel machine learning methods for solving Remote Sensing challenges, mostly based on applications of optimal transport to problems such as e.g. generalized domain adaptation, graph classification and prediction, or causality in remote sensing

CominLabs Dynalearn (2020–2024)

This project aims at contributing on the interactions between deep learning and physical models in a two-fold way: i) by exploring how dynamical formulation of learning process can help in understanding better learning deep neural architectures, as well as proposing new learning paradigms based on the regularization of the flows of information; ii) By leveraging on novel neural architectures and available data to devise new data-driven dynamical simulation models.

ANR PRC OATMIL (2017–2021)

The OATMIL project is funded by ANR. The goal is to bring innovative use of optimal transport in machine learning, and machine learning for the computation of optimal transport problems. I serve as coordinator for this project

Talks & Courses

Machine Learning Summer School in Saint-Etienne
VISUM 2020 DataScience summer school
NCCV - Keynote Speaker

Recent Publications

Quickly discover relevant content by filtering publications.
(2020). Learning with minibatch Wasserstein : asymptotic and gradient properties. the 23nd International Conference on Artificial Intelligence and Statistics.


(2019). Sliced Gromov-Wasserstein. NeurIPS 2019 - Thirty-third Conference on Neural Information Processing Systems.



  • (+33) 2 97 01 72 13
  • Campus de Tohannic, Vannes, Bretagne 56000
  • Enter ENSIBS Building and take the stairs to Office C103 on Floor 1