Paul Liautaud

Paul Liautaud

PhD candidate in Statistical & Machine Learning

Sorbonne University


I am a PhD student in Statistical/Machine Learning in the Statistics department of the LPSM (Sorbonne University, Paris) and in THOTH team (INRIA, Grenoble). My research activity is supervised by Pierre Gaillard & Olivier Wintenberger.

My main work focuses on sequential algorithms & online learning theory.

In addition to research, I am dedicated to teaching sessions in mathematics and computer science. If you are interested you can go to the teaching page.

  • Artificial Intelligence
  • Deep & Machine Learning
  • Online Learning
  • Stochastic Algorithms
  • MSc in Computer Science & Mathematics, 2022

    Sorbonne University

  • MSc in Mathematics & Applications, 2021

    Sorbonne University

  • BSc in Mathematics, 2020

    Sorbonne & Aix-Marseille University


PhD student
Oct 2022 – Present Paris
Topic: Online Boosting.
Stanislas College
Teaching & Examination
Mar 2022 – Present Paris
Mathematics and Python programing (MPSI, PSI, ECG).
Resarch Internship (Master Thesis)
Apr 2022 – Sep 2022 Paris
Topic: Online (Gradient) Boosting.
Resarch Internship
May 2021 – Aug 2021 Paris
Topic: Phase transition for the compressed sensing problem with Gaussian matrices and possibly with more structured random matrices.


Brownian Motion
Construction, properties & application of Brownian Motion. Co-worker T. Jaffard
Gradient Boosting
Gradient Boosting with Trees. Co-worker L. Ferraris
Training GANs with Visual Transformers. Co-workers L. LeBoudec, N. Olivain
Kernel Random Forest
Random Forests seen as Kernel methods. Co-worker L. Ferraris