Skip to main content
arXiv is now an independent nonprofit! Learn more

Showing 1–14 of 14 results for author: Azizian, W

Searching in archive cs. Search in all archives.
.
  1. arXiv:2510.01163  [pdf, ps, other

    cs.LG stat.ML

    How Does the Pretraining Distribution Shape In-Context Learning? A Fundamental Trade-Off

    Authors: Waïss Azizian, Ali Hasan

    Abstract: The factors driving the performance of in-context learning (ICL) in large language models (LLMs) remain poorly understood despite ICL's surprising effectiveness, enabling models to adapt to new tasks from only a handful of examples. To clarify and improve these capabilities, we characterize how the statistical properties of the pretraining distribution (e.g., tail behavior, coverage) shape ICL. We… ▽ More

    Submitted 24 June, 2026; v1 submitted 1 October, 2025; originally announced October 2025.

    Comments: 57 pages, 15 figures; to be presented at ICML 2026

    ACM Class: G.3; I.2.6

  2. arXiv:2506.08572  [pdf, ps, other

    cs.LG cs.AI cs.CL stat.ML

    The Geometries of Truth Are Orthogonal Across Tasks

    Authors: Waiss Azizian, Michael Kirchhof, Eugene Ndiaye, Louis Bethune, Michal Klein, Pierre Ablin, Marco Cuturi

    Abstract: Large Language Models (LLMs) have demonstrated impressive generalization capabilities across various tasks, but their claim to practical relevance is still mired by concerns on their reliability. Recent works have proposed examining the activations produced by an LLM at inference time to assess whether its answer to a question is correct. Some works claim that a "geometry of truth" can be learned… ▽ More

    Submitted 4 July, 2025; v1 submitted 10 June, 2025; originally announced June 2025.

  3. arXiv:2503.16398  [pdf, ps, other

    math.OC cs.LG

    The global convergence time of stochastic gradient descent in non-convex landscapes: Sharp estimates via large deviations

    Authors: Waïss Azizian, Franck Iutzeler, Jérôme Malick, Panayotis Mertikopoulos

    Abstract: In this paper, we examine the time it takes for stochastic gradient descent (SGD) to reach the global minimum of a general, non-convex loss function. We approach this question through the lens of randomly perturbed dynamical systems and large deviations theory, and we provide a tight characterization of the global convergence time of SGD via matching upper and lower bounds. These bounds are domina… ▽ More

    Submitted 7 June, 2025; v1 submitted 20 March, 2025; originally announced March 2025.

    Comments: 67 pages, 10 figures

    MSC Class: Primary 90C15; 90C26; 60F10; secondary 90C30; 68Q32

  4. arXiv:2410.21231  [pdf, ps, other

    cs.LG cs.MS math.OC

    $\texttt{skwdro}$: a library for Wasserstein distributionally robust machine learning

    Authors: Florian Vincent, Waïss Azizian, Franck Iutzeler, Jérôme Malick

    Abstract: We present skwdro, a Python library for training robust machine learning models. The library is based on distributionally robust optimization using Wasserstein distances, popular in optimal transport and machine learnings. The goal of the library is to make the training of robust models easier for a wide audience by proposing a wrapper for PyTorch modules, enabling model loss' robustification with… ▽ More

    Submitted 9 January, 2026; v1 submitted 28 October, 2024; originally announced October 2024.

    Comments: 7 pages 2 figures

    MSC Class: 90C17; 90C15 ACM Class: I.2.6; I.2.5; G.4; G.1.6

  5. arXiv:2406.09241  [pdf, ps, other

    math.OC cs.LG math.PR stat.ML

    What is the long-run distribution of stochastic gradient descent? A large deviations analysis

    Authors: Waïss Azizian, Franck Iutzeler, Jérôme Malick, Panayotis Mertikopoulos

    Abstract: In this paper, we examine the long-run distribution of stochastic gradient descent (SGD) in general, non-convex problems. Specifically, we seek to understand which regions of the problem's state space are more likely to be visited by SGD, and by how much. Using an approach based on the theory of large deviations and randomly perturbed dynamical systems, we show that the long-run distribution of SG… ▽ More

    Submitted 18 May, 2026; v1 submitted 13 June, 2024; originally announced June 2024.

    Comments: 71 pages, 3 figures; presented in ICML 2024

    MSC Class: Primary 90C15; 90C26; 60F10; secondary 90C30; 68Q32

  6. arXiv:2405.13592  [pdf, other

    cs.LG math.OC

    Almost sure convergence rates of stochastic gradient methods under gradient domination

    Authors: Simon Weissmann, Sara Klein, Waïss Azizian, Leif Döring

    Abstract: Stochastic gradient methods are among the most important algorithms in training machine learning problems. While classical assumptions such as strong convexity allow a simple analysis they are rarely satisfied in applications. In recent years, global and local gradient domination properties have shown to be a more realistic replacement of strong convexity. They were proved to hold in diverse setti… ▽ More

    Submitted 15 March, 2025; v1 submitted 22 May, 2024; originally announced May 2024.

  7. arXiv:2312.09860  [pdf, other

    cs.LG stat.CO

    Automatic Rao-Blackwellization for Sequential Monte Carlo with Belief Propagation

    Authors: Waïss Azizian, Guillaume Baudart, Marc Lelarge

    Abstract: Exact Bayesian inference on state-space models~(SSM) is in general untractable, and unfortunately, basic Sequential Monte Carlo~(SMC) methods do not yield correct approximations for complex models. In this paper, we propose a mixed inference algorithm that computes closed-form solutions using belief propagation as much as possible, and falls back to sampling-based SMC methods when exact computatio… ▽ More

    Submitted 15 December, 2023; originally announced December 2023.

  8. arXiv:2305.17076  [pdf, other

    cs.LG stat.ML

    Exact Generalization Guarantees for (Regularized) Wasserstein Distributionally Robust Models

    Authors: Waïss Azizian, Franck Iutzeler, Jérôme Malick

    Abstract: Wasserstein distributionally robust estimators have emerged as powerful models for prediction and decision-making under uncertainty. These estimators provide attractive generalization guarantees: the robust objective obtained from the training distribution is an exact upper bound on the true risk with high probability. However, existing guarantees either suffer from the curse of dimensionality, ar… ▽ More

    Submitted 6 November, 2023; v1 submitted 26 May, 2023; originally announced May 2023.

    Comments: 49 pages, 2 figures; to be presented at the 37th Annual Conference on Neural Information Processing Systems (NeurIPS 2023)

    Journal ref: 37th Annual Conference on Neural Information Processing Systems (NeurIPS 2023), Dec 2023, New Orleans, United States

  9. arXiv:2211.08043  [pdf, ps, other

    math.OC cs.LG

    The rate of convergence of Bregman proximal methods: Local geometry vs. regularity vs. sharpness

    Authors: Waïss Azizian, Franck Iutzeler, Jérôme Malick, Panayotis Mertikopoulos

    Abstract: We examine the last-iterate convergence rate of Bregman proximal methods - from mirror descent to mirror-prox and its optimistic variants - as a function of the local geometry induced by the prox-mapping defining the method. For generality, we focus on local solutions of constrained, non-monotone variational inequalities, and we show that the convergence rate of a given method depends sharply on i… ▽ More

    Submitted 12 December, 2024; v1 submitted 15 November, 2022; originally announced November 2022.

    Comments: 30 pages, 3 figures, 2 tables

    MSC Class: Primary 65K15; 90C33; secondary 68Q25; 68Q32

  10. arXiv:2107.01906  [pdf, ps, other

    math.OC cs.LG

    The Last-Iterate Convergence Rate of Optimistic Mirror Descent in Stochastic Variational Inequalities

    Authors: Waïss Azizian, Franck Iutzeler, Jérôme Malick, Panayotis Mertikopoulos

    Abstract: In this paper, we analyze the local convergence rate of optimistic mirror descent methods in stochastic variational inequalities, a class of optimization problems with important applications to learning theory and machine learning. Our analysis reveals an intricate relation between the algorithm's rate of convergence and the local geometry induced by the method's underlying Bregman function. We qu… ▽ More

    Submitted 5 July, 2021; originally announced July 2021.

    Comments: 31 pages, 3 figures, 1 table; to be presented at the 34th Annual Conference on Learning Theory (COLT 2021)

    MSC Class: 65K15; 90C33 (Primary) 68Q25; 68Q32 (Secondary)

  11. arXiv:2006.15646  [pdf, other

    cs.LG cs.DM stat.ML

    Expressive Power of Invariant and Equivariant Graph Neural Networks

    Authors: Waïss Azizian, Marc Lelarge

    Abstract: Various classes of Graph Neural Networks (GNN) have been proposed and shown to be successful in a wide range of applications with graph structured data. In this paper, we propose a theoretical framework able to compare the expressive power of these GNN architectures. The current universality theorems only apply to intractable classes of GNNs. Here, we prove the first approximation guarantees for p… ▽ More

    Submitted 6 June, 2021; v1 submitted 28 June, 2020; originally announced June 2020.

    Comments: Appears in: Proceedings of the 9th International Conference on Learning Representations, ICLR 2021. 39 pages

    ACM Class: G.1.6; I.2.6

  12. arXiv:2001.00602  [pdf, other

    cs.LG math.OC stat.ML

    Accelerating Smooth Games by Manipulating Spectral Shapes

    Authors: Waïss Azizian, Damien Scieur, Ioannis Mitliagkas, Simon Lacoste-Julien, Gauthier Gidel

    Abstract: We use matrix iteration theory to characterize acceleration in smooth games. We define the spectral shape of a family of games as the set containing all eigenvalues of the Jacobians of standard gradient dynamics in the family. Shapes restricted to the real line represent well-understood classes of problems, like minimization. Shapes spanning the complex plane capture the added numerical challenges… ▽ More

    Submitted 9 March, 2020; v1 submitted 2 January, 2020; originally announced January 2020.

    Comments: Appears in: Proceedings of the 23rd International Conference on Artificial Intelligence and Statistics (AISTATS 2020). 34 pages

    MSC Class: G.1.6; I.2.6 ACM Class: G.1.6; I.2.6

  13. arXiv:1906.07300  [pdf, ps, other

    cs.LG math.OC stat.ML

    Linear Lower Bounds and Conditioning of Differentiable Games

    Authors: Adam Ibrahim, Waïss Azizian, Gauthier Gidel, Ioannis Mitliagkas

    Abstract: Recent successes of game-theoretic formulations in ML have caused a resurgence of research interest in differentiable games. Overwhelmingly, that research focuses on methods and upper bounds on their speed of convergence. In this work, we approach the question of fundamental iteration complexity by providing lower bounds to complement the linear (i.e. geometric) upper bounds observed in the litera… ▽ More

    Submitted 15 September, 2020; v1 submitted 17 June, 2019; originally announced June 2019.

    Comments: ICML 2020 final version

    Journal ref: Proceedings of the 37 th International Conference on Machine Learning, Vienna, Austria, PMLR 119, 2020

  14. arXiv:1906.05945  [pdf, other

    cs.LG math.OC stat.ML

    A Tight and Unified Analysis of Gradient-Based Methods for a Whole Spectrum of Games

    Authors: Waïss Azizian, Ioannis Mitliagkas, Simon Lacoste-Julien, Gauthier Gidel

    Abstract: We consider differentiable games where the goal is to find a Nash equilibrium. The machine learning community has recently started using variants of the gradient method (GD). Prime examples are extragradient (EG), the optimistic gradient method (OG) and consensus optimization (CO), which enjoy linear convergence in cases like bilinear games, where the standard GD fails. The full benefits of theses… ▽ More

    Submitted 7 July, 2020; v1 submitted 13 June, 2019; originally announced June 2019.

    Comments: Appears in: Proceedings of the 23rd International Conference on Artificial Intelligence and Statistics (AISTATS 2020). 39 pages. Minor modification regarding prior work in comparison to the AISTATS Proceedings

    ACM Class: G.1.6; I.2.6