Publications

Publications by categories in reversed chronological order. See also Google Scholar.

* means equal contribution.

© 2024 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

preprints

  1. Multi-Source and Test-Time Domain Adaptation on Multivariate Signals using Spatio-Temporal Monge Alignment
    Théo Gnassounou*, Antoine Collas*, Rémi Flamary, Karim Lounici, and Alexandre Gramfort
    2024
    Submitted to Journal of Machine Learning Research (JMLR)
    arXiv
  2. SKADA-Bench: Benchmarking Unsupervised Domain Adaptation Methods with Realistic Validation
    Yanis Lalou*, Théo Gnassounou*, Antoine Collas*, Antoine Mathelin, Oleksii Kachaiev, Ambroise Odonnat, Alexandre Gramfort, Thomas Moreau, and Rémi Flamary
    2024
    Submitted to International Conference on Learning Representations (ICLR) 2025
    arXiv GitHub
  3. Geodesic Optimization for Predictive Shift Adaptation on EEG data
    Apolline Mellot*, Antoine Collas*, Sylvain Chevallier, Alexandre Gramfort, and Denis A Engemann
    2024
    Accepted Spotlight at NeurIPS 2024, Vancouver, Canada
    arXiv
  4. Weakly supervised covariance matrices alignment through Stiefel matrices estimation for MEG applications
    Antoine Collas, Rémi Flamary, and Alexandre Gramfort
    2024
    arXiv

softwares

  1. SKADA : Scikit Adaptation
    Théo Gnassounou, Oleksii Kachaiev, Rémi Flamary, Antoine Collas, Yanis Lalou, Antoine Mathelin, Alexandre Gramfort, Ruben Bueno, Florent Michel, Apolline Mellot, Virginie Loison, Ambroise Odonnat, and Thomas Moreau
    2024
    DOI GitHub doc

Theses

  1. Riemannian geometry for statistical estimation and learning: application to remote sensing
    Antoine Collas
    Ph. D. dissertation, Université Paris-Saclay, 2022
    HAL Slides

Book Chapters

  1. The Fisher-Rao geometry of CES distributions
    Florent Bouchard, Arnaud Breloy, Antoine Collas, Alexandre Renaux, and Guillaume Ginolhac
    In Springer (to appear), 2023
    arXiv

Journal Articles

  1. Harmonizing and aligning M/EEG datasets with covariance-based techniques to enhance predictive regression modeling
    Apolline Mellot, Antoine Collas, Pedro L. C. Rodrigues, Denis Engemann, and Alexandre Gramfort
    Imaging Neuroscience, 2023
    bioRxiv DOI GitHub
  2. Parametric information geometry with the package Geomstats
    Alice Le Brigant, Jules Deschamps, Antoine Collas, and Nina Miolane
    ACM Transactions on Mathematical Software, 2023
    arXiv DOI GitHub Slides
  3. Riemannian optimization for non-centered mixture of scaled Gaussian distributions
    Antoine Collas, Arnaud Breloy, Chengfang Ren, Guillaume Ginolhac, and Jean-Philippe Ovarlez
    IEEE Transactions on Signal Processing, 2023
    arXiv DOI GitHub
  4. Probabilistic PCA From Heteroscedastic Signals: Geometric Framework and Application to Clustering
    Antoine Collas, Florent Bouchard, Arnaud Breloy, Guillaume Ginolhac, Chengfang Ren, and Jean-Philippe Ovarlez
    IEEE Transactions on Signal Processing, 2021
    DOI GitHub PDF
  5. Robust Low-Rank Change Detection for Multivariate SAR Image Time Series
    Ammar Mian, Antoine Collas, Arnaud Breloy, Guillaume Ginolhac, and Jean-Philippe Ovarlez
    IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, 2020
    DOI GitHub PDF

Conference Articles

  1. Physics-informed and Unsupervised Riemannian Domain Adaptation for Machine Learning on Heterogeneous EEG Datasets
    Apolline Mellot, Antoine Collas, Sylvain Chevallier, Denis Engemann, and Alexandre Gramfort
    In 2024 32th European Signal Processing Conference (EUSIPCO), Lyon, France, 2024
    arXiv Poster
  2. Entropic Wasserstein Component Analysis
    Antoine Collas, Titouan Vayer, Rémi Flamary, and Arnaud Breloy
    In IEEE Machine Learning for Signal Processing (MLSP) - Rome, Italy, 2023
    arXiv GitHub doc Poster Slides
  3. Apprentissage robuste de distance par géométrie riemannienne
    Antoine Collas, Arnaud Breloy, Guillaume Ginolhac, Chengfang Ren, and Jean-Philippe Ovarlez
    In GRETSI 2022 XXVIIIème colloque, Nancy, France, 2022
    HAL GitHub Slides
  4. Robust Geometric Metric Learning
    Antoine Collas, Arnaud Breloy, Guillaume Ginolhac, Chengfang Ren, and Jean-Philippe Ovarlez
    In 2022 30th European Signal Processing Conference (EUSIPCO), Belgrade, Serbia, 2022
    Awarded arXiv GitHub Poster Slides
  5. On The Use of Geodesic Triangles Between Gaussian Distributions for Classification Problems
    Antoine Collas, Florent Bouchard, Guillaume Ginolhac, Arnaud Breloy, Chengfang Ren, and Jean-Philippe Ovarlez
    In ICASSP 2022 - 2022 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Singapore, 2022
    DOI GitHub PDF Poster Slides
  6. A Tyler-Type Estimator of Location and Scatter Leveraging Riemannian Optimization
    Antoine Collas, Florent Bouchard, Arnaud Breloy, Chengfang Ren, Guillaume Ginolhac, and Jean-Philippe Ovarlez
    In ICASSP 2021 - 2021 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), Toronto, Canada (Virtual), 2021
    DOI GitHub PDF Poster Slides