Léo Gautheron

Welcome to my professional web page.

Short bio

I am currently a Software Engineer at Synapse Défense working on web app development.

I was a post-doctoral researcher between December 9, 2020 and September 3rd, 2021 at the Hubert Curien Laboratory in the Data Intelligence team under the supervision of Pr. Marc Sebban, Pr. Amaury Habrard.

I defended my PhD and became a doctor in Computer Science on December 8, 2020. The work carried during my PhD was done at the Jean Monnet University of Saint-Etienne (France) since October 2, 2017. I worked under the supervision of Pr. Marc Sebban, Pr. Amaury Habrard and Dr. Emilie Morvant at the Hubert Curien Laboratory in the Data Intelligence team. My PhD was funded by a ministerial fellowship and its title is "Learning Tailored Data Representations from Few Labeled Examples". You can find the abstract of the thesis in the last page of the manuscript.

I received in 2017 a Master degree in Computer Science from the Machine Learning & Data Mining Master. During this Master, I did a research internship at CREATIS (a Biomedical Imaging Research Laboratory) under the supervision of Dr. Ievgen Redko and Pr. Carole Lartizien. Its subject was "Domain adaptation using Optimal Transport: application to prostate cancer mapping".

Research Interests:

Classification, Imbalanced Data, Representation Learning, Transfer Learning, Optimal Transport, Boosting, Kernel Methods.

Publications

  • MetaAP: A meta-tree-based ranking algorithm optimizing the average precision from imbalanced data.
    Rémi Viola, Léo Gautheron, Amaury Habrard and Marc Sebban
    Pattern Recognition Letters, volume 161, pages 161-167. 2022.
    [code][PDF]
  • POT: Python Optimal Transport
    Rémi Flamary, Nicolas Courty, Alexandre Gramfort, Mokhtar Z. Alaya, Aurélie Boisbunon, Stanislas Chambon, Laetitia Chapel, Adrien Corenflos, Kilian Fatras, Nemo Fournier, Léo Gautheron, Nathalie T.H. Gayraud, Hicham Janati, Alain Rakotomamonjy, Ievgen Redko, Antoine Rolet, Antony Schutz, Vivien Seguy, Danica J. Sutherland, Romain Tavenard, Alexander Tong, Titouan Vayer
    Journal of Machine Learning Research, volume 22, number 78, pages 1-8. 2021.
    [code][abs][PDF]
  • Learning Tailored Data Representations from Few Labeled Examples
    Léo Gautheron
    PhD thesis, 2020, Jean Monnet University, France.
    [PDF][slides]
  • Landmark-based Ensemble Learning with Random Fourier Features and Gradient Boosting.
    Léo Gautheron, Pascal Germain, Amaury Habrard, Guillaume Metzler, Emilie Morvant, Marc Sebban and Valentina Zantedeschi
    European Conference on Machine Learning & Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD), 2020, On-line.
    [code][PDF (Extended version as Chapter 3 in my PhD manuscript)]
  • Apprentissage d'ensemble basée sur des points de repère avec des caractéristiques de Fourier aléatoires et un renforcement du gradient.
    Léo Gautheron, Pascal Germain, Amaury Habrard, Guillaume Metzler, Emilie Morvant, Marc Sebban and Valentina Zantedeschi
    Conférence sur l'Apprentissage automatique (CAp 2020), 2020, On-line.
    [code][video][PDF (Extended version as Chapter 3 in my PhD manuscript)]
  • Metric Learning from Imbalanced Data with Generalization Guarantees
    Léo Gautheron, Emilie Morvant, Amaury Habrard, Marc Sebban
    Pattern Recognition Letters, volume 133, pages 298-304. 2020.
    [code][PDF (Extended version as Chapter 2 in my PhD manuscript)]
  • Metric Learning from Imbalanced Data
    Léo Gautheron, Emilie Morvant, Amaury Habrard, Marc Sebban
    IEEE International Conference on Tools with Artificial Intelligence (ICTAI), 2019, Portland, United States.
    [PDF][code][PDF (Extended version as Chapter 2 in my PhD manuscript)]
  • Revisite des "random Fourier features" basée sur l'apprentissage PAC-Bayésien via des points d'intérêts
    Léo Gautheron, Pascal Germain, Amaury Habrard, Gaël Letarte, Emilie Morvant, Marc Sebban, Valentina Zantedeschi
    Conférence sur l'Apprentissage automatique (CAp 2019), 2019, Toulouse, France.
    [PDF (Extended version as Chapter 3 in my PhD manuscript)]
  • Feature Selection for Unsupervised Domain Adaptation using Optimal Transport
    Léo Gautheron, Ievgen Redko, Carole Lartizien
    European Conference on Machine Learning & Principles and Practice of Knowledge Discovery in Databases (ECML-PKDD), 2018, Dublin, Ireland.
    [PDF][code][Poster][PDF (Extended version as Chapter 4 in my PhD manuscript)]
  • Apprentissage de métrique pour la classification supervisée de données déséquilibrées
    Léo Gautheron, Amaury Habrard, Emilie Morvant, Marc Sebban
    Conférence sur l'Apprentissage automatique (CAp 2018), 2018, Rouen, France.
    [PDF (Extended version as Chapter 2 in my PhD manuscript)]
  • Adaptation de domaine pour la détection automatique du cancer de la prostate en imagerie IRM multiparamétrique
    Léo Gautheron, Ievgen Redko, Carole Lartizien
    XXVIème colloque GRETSI (GRETSI 2017), 2017, Juan-Les-Pins, France.
    [PDF]
  • Domain adaptation using Optimal Transport: application to prostate cancer mapping
    Léo Gautheron
    Master’s thesis, 2017, Jean Monnet University, France.
    [PDF]

Code & Software

The code associated with a publication can be found in the Publications section.
  • 2019 - Sudoku [link]
  • 2017 - A domain adaptation benchmark in Python3 [link]
  • 2017 - A web application in JavaScript for hand written digit recognition. [link]