publications

publications by categories in reversed chronological order. generated by jekyll-scholar.

2025

  1. preprint
    The Computational Advantage of Depth: Learning High-Dimensional Hierarchical Functions with Gradient Descent
    Yatin Dandi, Luca Pesce, Lenka Zdeborová, and Florent Krzakala
    2025
    arXiv
  2. AISTATS
    A Random Matrix Theory Perspective on the Spectrum of Learned Features and Asymptotic Generalization Capabilities
    Yatin Dandi, Luca Pesce, Hugo Cui, Florent Krzakala, Yue M. Lu, and Bruno Loureiro
    In Artificial Intelligence and Statistics, 2025
    arXiv

2024

  1. preprint
    Repetita iuvant: Data repetition allows sgd to learn high-dimensional multi-index functions
    Luca Arnaboldi, Yatin Dandi, Florent Krzakala, Luca Pesce, and Ludovic Stephan
    2024
    arXiv
  2. ICML
    Online Learning and Information Exponents: On The Importance of Batch size, and Time/Complexity Tradeoffs
    Luca Arnaboldi, Yatin Dandi, Florent Krzakala, Bruno Loureiro, Luca Pesce, and Ludovic Stephan
    In Proceedings of the 41th International Conference on Machine Learning, 2024
    arXiv PDF
  3. ICML
    Asymptotics of feature learning in two-layer networks after one gradient-step
    Hugo Cui, Luca Pesce, Yatin Dandi, Florent Krzakala, Yue M. Lu, Lenka Zdeborová, and Bruno Loureiro
    In Proceedings of the 41th International Conference on Machine Learning, 2024
    arXiv PDF
  4. ICML
    The benefits of reusing batches for gradient descent in two-layer networks: Breaking the curse of information and leap exponents
    Yatin Dandi, Emanuele Troiani, Luca Arnaboldi, Luca Pesce, Lenka Zdeborová, and Florent Krzakala
    In Proceedings of the 41th International Conference on Machine Learning, 2024
    arXiv PDF
  5. JMLR
    Learning Two-Layer Neural Networks, One (Giant) Step at a Time
    Yatin Dandi, Florent Krzakala, Bruno Loureiro, Luca Pesce, and Ludovic Stephan
    Journal of Machine Learning Research, 2024
    arXiv
  6. JSTAT
    Theory and applications of the Sum-Of-Squares technique
    Francis Bach, Elisabetta Cornacchia, Luca Pesce, and Giovanni Piccioli
    Journal of Statistical Mechanics: Theory and Experiment, 2024
    arXiv PDF

2023

  1. ICML
    Are Gaussian Data All You Need? The Extents and Limits of Universality in High-Dimensional Generalized Linear Estimation
    Luca Pesce, Florent Krzakala, Bruno Loureiro, and Ludovic Stephan
    In Proceedings of the 40th International Conference on Machine Learning, 2023
    arXiv PDF

2022

  1. NeurIPS
    Subspace clustering in high-dimensions: Phase transitions & Statistical-to-Computational gap
    Luca Pesce, Bruno Loureiro, Florent Krzakala, and Lenka Zdeborová
    In Advances in Neural Information Processing Systems, 2022
    arXiv PDF