Florent Gbelidji

Florent Gbelidji

Paris, Île-de-France, France
2 k abonnés + de 500 relations

Devenir membre pour voir le profil

Activité

S’inscrire pour voir toute l’activité

Expérience

  • Graphique Hugging Face

    Hugging Face

    Paris, Île-de-France, France

  • -

  • -

    Paris

  • -

    Paris, Île-de-France, France

  • -

    Vancouver, Canada Area

  • -

    Princeton, New Jersey, USA

  • -

    Paris

  • -

    Gif-sur-Yvette, France

  • -

    Ploiesti, Romania

Formation

  • Graphique CentraleSupélec

    CentraleSupélec

    -

    Activités et associations :President of the Student Sports Union, Member of the rugby team.

    Top-ranked French School of Engineering and Applied Science (SEAS) working towards a Master of Science (MSc)

    • Expected graduation date : September 2019,

  • -

    Master of Science in Mathematics, Modeling and Machine Learning.

    Key subjects studied: Machine Learning, Optimisation, Stochastic Algorithms, Computer Vision, Inverse Problems, Nonparametric Statistics

  • -

    Activités et associations :President of the Student Union

    • Preparatory classes (Mathematics and Physics section) for the competitive national entrance
    examinations to gain entrance into France’s top graduate Schools.

    • Key subjects studied : Algebra, Analysis, General Mechanics, Optics, Thermodynamics, Electromagnetism, Fluid Mechanics, General Chemistry, Organic Chemistry, Philosophy.

  • -

Publications

  • Deep transform networks for scalable learning of MR reconstruction

    Proc. Intl. Soc. Mag. Reson. Med. 27 (2019)

    In this work we introduce RadixNet, a fast, scalable, transform network architecture based on the Cooley-Tukey FFT, and use it in a fully-learnt iterative reconstruction with a residual dense U-Net image regularization. Results show that fast transform networks can be trained at 256x256 dimensions and outperform the FFT.

    Other authors
    See publication

Brevets

  • Machine-Learned Network for Fourier Transform in Reconstruction for Medical Imaging

    Émis le US US20190378311

    For low-complexity to learned reconstruction and/or learned Fourier transform-based operators for reconstruction, a neural network is used for the transform operators. The network architecture is modeled on the Cooley-Tukey fast Fourier transform (FFT) approach. By splitting input data before recursive calls in the network architecture, the network may be trained to perform the transform with similar complexity as FFT. The learned operators may be used in a trained network for reconstruction…

    For low-complexity to learned reconstruction and/or learned Fourier transform-based operators for reconstruction, a neural network is used for the transform operators. The network architecture is modeled on the Cooley-Tukey fast Fourier transform (FFT) approach. By splitting input data before recursive calls in the network architecture, the network may be trained to perform the transform with similar complexity as FFT. The learned operators may be used in a trained network for reconstruction, such as with a learned iterative framework and image regularizer.

    Other inventors
    • Boris Mailhé
    • Simon Arberet
    See patent

Langues

  • Anglais

    Capacité professionnelle générale

  • Français

    Bilingue ou langue natale

  • Espagnol

    Compétence professionnelle limitée

  • Russe

    Notions

Plus d’activités de Florent

Voir le profil complet de Florent

  • Découvrir vos relations en commun
  • Être mis en relation
  • Contacter Florent directement
Devenir membre pour voir le profil complet

Autres profils similaires

Ajoutez de nouvelles compétences en suivant ces cours

Voir tous les cours