Ioannis Mitliagkas

Biography

Ioannis Mitliagkas is an associate professor in the Department of Computer Science and Operations Research (DIRO) at Université de Montréal, as well as a Core Academic member of Mila – Quebec Artificial Intelligence Institute and a Canada CIFAR AI Chair. I hold a part-time position as a staff research scientist at Google DeepMind Montréal.

Previously, I was a postdoctoral scholar in the departments of statistics and computer science at Stanford University. I obtained my PhD from the Department of Electrical and Computer Engineering at the University of Texas at Austin.

My research interests lie in statistical learning and inference, with a focus on optimization, efficient large-scale and distributed algorithms, statistical learning theory and MCMC methods. My recent research has focused on methods for efficient and adaptive optimization, understanding the interaction between optimization and the dynamics of large-scale learning systems, and the dynamics of games.

Current Students

Mehdi Inane Ahmed

PhD - Université de Montréal

reyhane.askari.hemmat@mila.quebec

Ange-Clément Akazan

Research Intern - Université de Montréal

Reyhane Askari Hemmat

PhD - Université de Montréal

Co-supervisor :

Nicolas Le Roux

Oscar Clivio

Research Intern - Université de Montréal

Co-supervisor :

PhD - Université de Montréal

Kilian Fatras

Postdoctorate - McGill University

Principal supervisor :

Charles Guille-escuret

PhD - Université de Montréal

Mehrab Hamidi

PhD - Université de Montréal

Adam Ibrahim

PhD - Université de Montréal

Co-supervisor :

Irina Rish

Zichu Liu

PhD - Université de Montréal

Principal supervisor :

Divyat Mahajan

PhD - Université de Montréal

Mehrnaz Mofakhami

Master's Research - Université de Montréal

Principal supervisor :

PhD - Université de Montréal

Vincent Quirion

Master's Research - Université de Montréal

Additive Decoders for Latent Variables Identification and Cartesian-Product Extrapolation

Blog Posts

March 18, 2024

Sébastien Lachapelle

Divyat Mahajan

Ioannis Mitliagkas

Simon Lacoste-Julien

Read the article

Publications

Adversarial score matching and improved sampling for image generation

Alexia Jolicoeur-Martineau

Rémi Piché-Taillefer

Remi Tachet des Combes

Denoising Score Matching with Annealed Langevin Sampling (DSM-ALS) has recently found success in generative modeling. The approach works by … (see more)first training a neural network to estimate the score of a distribution, and then using Langevin dynamics to sample from the data distribution assumed by the score network. Despite the convincing visual quality of samples, this method appears to perform worse than Generative Adversarial Networks (GANs) under the Fréchet Inception Distance, a standard metric for generative models. We show that this apparent gap vanishes when denoising the final Langevin samples using the score network. In addition, we propose two improvements to DSM-ALS: 1) Consistent Annealed Sampling as a more stable alternative to Annealed Langevin Sampling, and 2) a hybrid training formulation, composed of both Denoising Score Matching and adversarial objectives. By combining these two techniques and exploring different network architectures, we elevate score matching methods and obtain results competitive with state-of-the-art image generation on CIFAR-10.

2021-01-12

ICLR.cc/2021/Conference (poster)

openreview.net

Invariance Principle Meets Information Bottleneck for Out-of-Distribution Generalization

Kartik Ahuja

Ethan Caballero

Dinghuai Zhang

Jean-Christophe Gagnon-Audet

Yoshua Bengio

Irina Rish

The invariance principle from causality is at the heart of notable approaches such as invariant risk minimization (IRM) that seek to address… (see more) out-of-distribution (OOD) generalization failures. Despite the promising theory, invariance principle-based approaches fail in common classification tasks, where invariant (causal) features capture all the information about the label. Are these failures due to the methods failing to capture the invariance? Or is the invariance principle itself insufficient? To answer these questions, we revisit the fundamental assumptions in linear regression tasks, where invariance-based approaches were shown to provably generalize OOD. In contrast to the linear regression tasks, we show that for linear classification tasks we need much stronger restrictions on the distribution shifts, or otherwise OOD generalization is impossible. Furthermore, even with appropriate restrictions on distribution shifts in place, we show that the invariance principle alone is insufficient. We prove that a form of the information bottleneck constraint along with invariance helps address key failures when invariant features capture all the information about the label and also retains the existing success when they do not. We propose an approach that incorporates both of these principles and demonstrate its effectiveness in several experiments.

openreview.net

A Study of Condition Numbers for First-Order Optimization

Charles Guille-Escuret

Baptiste Goujaud

Manuela Girotti

The study of first-order optimization algorithms (FOA) typically starts with assumptions on the objective functions, most commonly smoothnes… (see more)s and strong convexity. These metrics are used to tune the hyperparameters of FOA. We introduce a class of perturbations quantified via a new norm, called *-norm. We show that adding a small perturbation to the objective function has an equivalently small impact on the behavior of any FOA, which suggests that it should have a minor impact on the tuning of the algorithm. However, we show that smoothness and strong convexity can be heavily impacted by arbitrarily small perturbations, leading to excessively conservative tunings and convergence issues. In view of these observations, we propose a notion of continuity of the metrics, which is essential for a robust tuning strategy. Since smoothness and strong convexity are not continuous, we propose a comprehensive study of existing alternative metrics which we prove to be continuous. We describe their mutual relations and provide their guaranteed convergence rates for the Gradient Descent algorithm accordingly tuned. Finally we discuss how our work impacts the theoretical understanding of FOA and their performances.

2020-12-10

ArXiv (preprint)

Linear Lower Bounds and Conditioning of Differentiable Games

Adam Ibrahim

Waiss Azizian

Recent successes of game-theoretic formulations in ML have caused a resurgence of research interest in differentiable games. Overwhelmingly,… (see more) 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 literature on a wide class of problems. We cast saddle-point and min-max problems as 2-player games. We leverage tools from single-objective convex optimisation to propose new linear lower bounds for convex-concave games. Notably, we give a linear lower bound for

2020-11-21

Proceedings of the 37th International Conference on Machine Learning (published)

Accelerating Smooth Games by Manipulating Spectral Shapes

Waiss Azizian

Damien Scieur

2020-06-03

Proceedings of the Twenty Third International Conference on Artificial Intelligence and Statistics (published)

Accelerating Smooth Games by Manipulating Spectral Shapes

Waiss Azizian

Damien Scieur

We use matrix iteration theory to characterize acceleration in smooth games. We define the spectral shape of a family of games as the set co… (see more)ntaining 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 in solving smooth games. In this framework, we describe gradient-based methods, such as extragradient, as transformations on the spectral shape. Using this perspective, we propose an optimal algorithm for bilinear games. For smooth and strongly monotone operators, we identify a continuum between convex minimization, where acceleration is possible using Polyak's momentum, and the worst case where gradient descent is optimal. Finally, going beyond first-order methods, we propose an accelerated version of consensus optimization.

2020-01-02

ArXiv (preprint)

Gintare Karolina Dziugaite

In Search of Robust Measures of Generalization

Alexandre Drouin

Brady Neal

Nitarshan Rajkumar

Ethan Caballero

Linbo Wang

Daniel M. Roy

One of the principal scientific challenges in deep learning is explaining generalization, i.e., why the particular way the community now tra… (see more)ins networks to achieve small training error also leads to small error on held-out data from the same population. It is widely appreciated that some worst-case theories -- such as those based on the VC dimension of the class of predictors induced by modern neural network architectures -- are unable to explain empirical performance. A large volume of work aims to close this gap, primarily by developing bounds on generalization error, optimization error, and excess risk. When evaluated empirically, however, most of these bounds are numerically vacuous. Focusing on generalization bounds, this work addresses the question of how to evaluate such bounds empirically. Jiang et al. (2020) recently described a large-scale empirical study aimed at uncovering potential causal relationships between bounds/measures and generalization. Building on their study, we highlight where their proposed methods can obscure failures and successes of generalization measures in explaining generalization. We argue that generalization measures should instead be evaluated within the framework of distributional robustness.

Stochastic Hamiltonian Gradient Methods for Smooth Games

Nicolas Loizou

Hugo Berard

Alexia Jolicoeur-Martineau

Pascal Vincent

The success of adversarial formulations in machine learning has brought renewed motivation for smooth games. In this work, we focus on the c… (see more)lass of stochastic Hamiltonian methods and provide the first convergence guarantees for certain classes of stochastic smooth games. We propose a novel unbiased estimator for the stochastic Hamiltonian gradient descent (SHGD) and highlight its benefits. Using tools from the optimization literature we show that SHGD converges linearly to the neighbourhood of a stationary point. To guarantee convergence to the exact solution, we analyze SHGD with a decreasing step-size and we also present the first stochastic variance reduced Hamiltonian method. Our results provide the first global non-asymptotic last-iterate convergence guarantees for the class of stochastic unconstrained bilinear games and for the more general class of stochastic games that satisfy a "sufficiently bilinear" condition, notably including some non-convex non-concave problems. We supplement our analysis with experiments on stochastic bilinear and sufficiently bilinear games, where our theory is shown to be tight, and on simple adversarial machine learning formulations.

2020-01-01

ICML (published)

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

Waiss Azizian

We consider diﬀerentiable games where the goal is to ﬁnd a Nash equilibrium. The machine learning community has recently started using v… (see more)ariants 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 bene-ﬁts of theses relatively new methods are not known as there is no uniﬁed analysis for both strongly monotone and bilinear games. We provide new analyses of the EG ’s local and global convergence properties and use is to get a tighter global convergence rate for OG and CO . Our analysis covers the whole range of settings between bilinear and strongly monotone games. It reveals that these methods converges via diﬀerent mechanisms at these extremes; in between, it exploits the most favorable mechanism for the given problem. We then prove that EG achieves the optimal rate for a wide class of algorithms with any number of extrapolations. Our tight analysis of EG ’s convergence rate in games shows that, unlike in convex minimization, EG may be much faster than GD .

2020-01-01

International Conference on Artificial Intelligence and Statistics (published)

dblp.uni-trier.de

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

Waiss Azizian

2020-01-01

International Conference on Artificial Intelligence and Statistics (published)

Adversarial target-invariant representation learning for domain generalization

Isabela Albuquerque

Joao Monteiro

Tiago Falk

In many applications of machine learning, the training and test set data come from different distributions, or domains. A number of domain g… (see more)eneralization strategies have been introduced with the goal of achieving good performance on out-of-distribution data. In this paper, we propose an adversarial approach to the problem. We propose a process that enforces pair-wise domain invariance while training a feature extractor over a diverse set of domains. We show that this process ensures invariance to any distribution that can be expressed as a mixture of the training domains. Following this insight, we then introduce an adversarial approach in which pair-wise divergences are estimated and minimized. Experiments on two domain generalization benchmarks for object recognition (i.e., PACS and VLCS) show that the proposed method yields higher average accuracy on the target domains in comparison to previously introduced adversarial strategies, as well as recently proposed methods based on learning invariant representations.

2019-11-03

arXiv.org (preprint)

dblp.uni-trier.de

Generalizing to unseen domains via distribution matching

Isabela Albuquerque

Joao Monteiro

Mohammad Javad Darvishi Bayazi

Tiago Falk

Supervised learning results typically rely on assumptions of i.i.d. data. Unfortunately, those assumptions are commonly violated in practice… (see more). In this work, we tackle this problem by focusing on domain generalization: a formalization where the data generating process at test time may yield samples from never-before-seen domains (distributions). Our work relies on a simple lemma: by minimizing a notion of discrepancy between all pairs from a set of given domains, we also minimize the discrepancy between any pairs of mixtures of domains. Using this result, we derive a generalization bound for our setting. We then show that low risk over unseen domains can be achieved by representing the data in a space where (i) the training distributions are indistinguishable, and (ii) relevant information for the task at hand is preserved. Minimizing the terms in our bound yields an adversarial formulation which estimates and minimizes pairwise discrepancies. We validate our proposed strategy on standard domain generalization benchmarks, outperforming a number of recently introduced methods. Notably, we tackle a real-world application where the underlying data corresponds to multi-channel electroencephalography time series from different subjects, each considered as a distinct domain.

2019-11-03

ArXiv (preprint)