1. Escaping mediocrity: how two-layer networks learn hard single-index models with SGD
    Luca ArnaboldiFlorent KrzakalaBruno Loureiro and Ludovic Stephan
    May 2023
  2. Learning Two-Layer Neural Networks, One (Giant) Step at a Time
    Yatin DandiFlorent KrzakalaBruno Loureiro, Luca Pesce and Ludovic Stephan
    May 2023
  3. A non-backtracking method for long matrix and tensor completion
    Ludovic Stephan and Yizhe Zhu
    Apr 2023
  4. Are Gaussian data all you need? Extents and limits of universality in high-dimensional generalized linear estimation
    Luca Pesce, Florent KrzakalaBruno Loureiro and Ludovic Stephan
    Feb 2023
  5. From high-dimensional & mean-field dynamics to dimensionless ODEs: A unifying approach to SGD in two-layers networks
    Luca ArnaboldiLudovic StephanFlorent Krzakala and Bruno Loureiro
    Feb 2023
  6. Universality laws for Gaussian mixtures in generalized linear models
    Yatin DandiLudovic StephanFlorent KrzakalaBruno Loureiro and Lenka Zdeborová
    Feb 2023


  1. Non-backtracking spectra of weighted inhomogeneous random graphs
    Ludovic Stephan and Laurent Massoulié
    Mathematical Statistics and Learning, Dec 2022
  2. Sparse random hypergraphs: Non-backtracking spectra and community detection
    Ludovic Stephan and Yizhe Zhu
    In 2022 IEEE 63rd Annual Symposium on Foundations of Computer Science (FOCS), Oct 2022
  3. Gaussian Universality of Linear Classifiers with Random Labels in High-Dimension
    Federica Gerace, Florent KrzakalaBruno LoureiroLudovic Stephan and Lenka Zdeborová
    May 2022
  4. Phase diagram of Stochastic Gradient Descent in high-dimensional two-layer neural networks
    Rodrigo Veiga, Ludovic StephanBruno LoureiroFlorent Krzakala and Lenka Zdeborová
    In Advances in Neural Information Processing Systems, May 2022


  1. A simpler spectral approach for clustering in directed networks
    Simon Coste and Ludovic Stephan
    Feb 2021


  1. Planting trees in graphs, and finding them back
    Laurent MassouliéLudovic Stephan and Don Towsley
    In Conference on Learning Theory, Jun 2019
  2. Robustness of Spectral Methods for Community Detection
    Ludovic Stephan and Laurent Massoulié
    In Conference on Learning Theory, Jun 2019