Waïss Azizian

waiss-profile-square.jpg

Paris, France

waiss.azizian@gmail.com

I am an ML Researcher at Morgan Stanley ML Research in Paris, which I joined in July 2026. My work there builds on my research on the optimization and reliability of modern machine learning systems.

Before that, I completed my PhD in machine learning and optimization at the LJK lab within Université Grenoble Alpes, defended in June 2026. I was fortunate to be advised by Franck Iutzeler, Jérôme Malick, and Panayotis Mertikopoulos. My committee was made of Francis Bach and John Duchi as reviewers, and Niao He, Anatoli Juditsky, Julien Mairal and Gabriel Peyré as examiners, whom I thank again for accepting to be part of it. Before starting my PhD, I studied at ENS Paris and graduated from the MVA master.

The aim of my research is two-fold: (i) advancing our understanding of the intricate phenomena at play in deep learning, using tools from optimization, dynamical systems, probability and statistics; (ii) leveraging this knowledge to deliver more reliable and efficient machine learning systems.

Keywords: stochastic optimization, deep learning, reliable ML, LLMs

Resume: PDF • webpage

contact

research

My research spans four main areas that contribute to a principled understanding of deep learning systems and their optimization dynamics. You can find a list of my publications on the publications page. Here are some of my main research projects:

You can browse my research activity on arXiv, Google Scholar, DBLP, GitHub, and LinkedIn.

news

Jul 23, 2026 I joined the Morgan Stanley ML Research team in Paris as an ML Researcher!
Jun 16, 2026 I defended my PhD thesis, “Stochastic Optimization Algorithms in Machine Learning: Dynamics, Convergence, Generalization”, at Université Grenoble Alpes (slides).
Jun 08, 2026 Our paper from my Morgan Stanley ML Research internship, “How Does the Pretraining Distribution Shape In-Context Learning? A Fundamental Trade-Off”, was accepted at ICML 2026! (paper).
May 13, 2026 I will be presenting our work on the long-run distribution of SGD at MaLGa, Università di Genova on May 13, 2026 (slides).
Apr 21, 2026 Had the pleasure of presenting our work on the long-run distribution of SGD at the ENS Lyon Machine Learning and Signal Processing seminar.

publications

2026

  1. How Does the Pretraining Distribution Shape In-Context Learning? A Fundamental Trade-Off
    Waïss Azizian and Ali Hasan
    In ICML, 2026
    arXiv

2025

  1. The geometries of truth are orthogonal across tasks
    Waïss Azizian, Michael Kirchhof, Eugene Ndiaye, and 4 more authors
    In ICML 2025 Workshop on Reliable and Responsible Foundation Models, 2025
    arXiv
  2. The global convergence time of stochastic gradient descent in non-convex landscapes: sharp estimates via large deviations
    Waïss Azizian, Franck Iutzeler, Jerome Malick, and 1 more author
    In ICML, 2025
    arXiv Poster
  3. Almost sure convergence of stochastic gradient methods under gradient domination
    Simon Weissmann, Sara Klein, Waïss Azizian, and 1 more author
    Transactions on Machine Learning Research, 2025
    arXiv

2024

  1. The rate of convergence of bregman proximal methods: local geometry versus regularity versus sharpness
    Waı̈ss Azizian, Franck Iutzeler, Jérôme Malick, and 1 more author
    SIAM Journal on Optimization, 2024
    arXiv Slides
  2. What is the long-run distribution of stochastic gradient descent? A large deviations analysis
    Waïss Azizian, Franck Iutzeler, Jerome Malick, and 1 more author
    In ICML, 2024
    arXiv Poster
  3. skwdro: a library for Wasserstein distributionally robust machine learning
    Florian Vincent, Waïss Azizian, Franck Iutzeler, and 1 more author
    arXiv: 2410.21231, 2024
    arXiv

2023

  1. Regularization for Wasserstein distributionally robust optimization
    Waïss Azizian, Franck Iutzeler, and Jérôme Malick
    ESAIM: Control, Optimisation and Calculus of Variations, 2023
    arXiv
  2. Exact generalization guarantees for (regularized) Wasserstein distributionally robust models
    Waïss Azizian, Franck Iutzeler, and Jérôme Malick
    In NeurIPS, 2023
    arXiv Poster Slides
  3. Automatic Rao-Blackwellization for sequential Monte Carlo with belief propagation
    Waïss Azizian, Guillaume Baudart, and Marc Lelarge
    In ICML 2023 Workshop on Structured Probabilistic Inference & Generative Modeling, 2023
    arXiv

2021

  1. Expressive power of invariant and equivariant graph neural networks
    Waiss Azizian and Marc Lelarge
    In ICLR , 2021
    arXiv Slides
  2. The last-iterate convergence rate of optimistic mirror descent in stochastic variational inequalities
    Waïss Azizian, Franck Iutzeler, Jérôme Malick, and 1 more author
    In COLT, 2021
    arXiv Poster Slides

2020

  1. Accelerating smooth games by manipulating spectral shapes
    Waïss Azizian, Damien Scieur, Ioannis Mitliagkas, and 2 more authors
    In AISTATS, 2020
    arXiv
  2. A tight and unified analysis of gradient-based methods for a whole spectrum of differentiable games
    Waïss Azizian, Ioannis Mitliagkas, Simon Lacoste-Julien, and 1 more author
    In AISTATS, 2020
    arXiv Slides
  3. Linear lower bounds and conditioning of differentiable games
    Adam Ibrahim, Waïss Azizian, Gauthier Gidel, and 1 more author
    In ICML, 2020
    arXiv