Portrait de Gauthier Gidel

Gauthier Gidel

Membre académique principal
gidelgau@mila.quebec
Chaire en IA Canada-CIFAR
Professeur adjoint, Université de Montréal, Département d'informatique et de recherche opérationnelle
Site web
Google Scholar

Biographie

Je suis professeur adjoint au Département d’informatique et de recherche opérationnelle (DIRO) de l'Université de Montréal et membre académique principal de Mila – Institut québécois d’intelligence artificielle. J'ai obtenu une bourse Borealis AI destinée aux étudiant·e·s des cycles supérieurs et je suis actuellement titulaire d'une chaire en IA Canada-CIFAR. J'ai travaillé chez DeepMind et Element AI, et j'ai récemment été un visiteur de longue durée au Simons Institute de l’Université de Californie à Berkeley. Mes intérêts de recherche se situent à l'intersection de la théorie des jeux, de l'optimisation et de l'apprentissage automatique.

Étudiants actuels

Rohan Banerjee
Maîtrise recherche - Polytechnique
Superviseur⋅e principal⋅e :
Julien Cohen-Adad
rohan.banerjee@mila.quebec
Site web
Github
Google Scholar
Quentin Bertrand
Postdoctorat - UdeM
Co-superviseur⋅e :
Simon Lacoste-Julien
quentin.bertrand@mila.quebec
Site web
Github
Google Scholar
Bilun Bi Lun Sun
Collaborateur·rice de recherche
bilun.sun@mila.quebec
Github
Joey Bose
Stagiaire de recherche - McGill
bosejoey@mila.quebec
Site web
Github
Google Scholar
David Dobre
Doctorat - UdeM
david-a.dobre@mila.quebec
Damien Ferbach
Doctorat - UdeM
damien.ferbach@mila.quebec
Google Scholar
Florian Golemo
Postdoctorat - McGill
Superviseur⋅e principal⋅e :
Derek Nowrouzezahrai
golemofl@mila.quebec
Site web
Github
Google Scholar
Marco Jiralerspong
Doctorat - UdeM
marco.jiralerspong@mila.quebec
Site web
Github
Google Scholar
Zichu Liu
Doctorat - UdeM
Co-superviseur⋅e :
Ioannis Mitliagkas
zichu.liu@mila.quebec
Andjela Mladenovic
Doctorat - UdeM
andjela.mladenovic@mila.quebec
Site web
Google Scholar
Mehrnaz Mofakhami
Maîtrise recherche - UdeM
Co-superviseur⋅e :
Ioannis Mitliagkas
mehrnaz.mofakhami@mila.quebec
Site web
Github
Google Scholar
Armin Moradi
Maîtrise recherche - UdeM
Superviseur⋅e principal⋅e :
Golnoosh Farnadi
armin.moradi@mila.quebec
Site web
Github
Google Scholar
Michael Noukhovitch
Doctorat - UdeM
Superviseur⋅e principal⋅e :
Aaron Courville
noukhovm@mila.quebec
Site web
Github
Google Scholar
Leo Schwinn
Visiteur de recherche indépendant - Technical Univeristy of Munich
leo.schwinn@mila.quebec
Site web
Github
Google Scholar
Danilo Vucetic
Doctorat - UdeM
danilo.vucetic@mila.quebec
Sophie Xhonneux
Doctorat - UdeM
Superviseur⋅e principal⋅e :
Jian Tang
sophie.xhonneux@mila.quebec

Publications

A Tight and Unified Analysis of Gradient-Based Methods for a Whole Spectrum of Differentiable Games
Waiss Azizian
Ioannis Mitliagkas
Simon Lacoste-Julien
Gauthier Gidel
We consider differentiable games where the goal is to find a Nash equilibrium. The machine learning community has recently started using v… (voir plus)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
Negative Momentum for Improved Game Dynamics
Gauthier Gidel
Reyhane Askari Hemmat
Mohammad Pezeshki
Gabriel Huang
Rémi LE PRIOL
Simon Lacoste-Julien
Ioannis Mitliagkas
Games generalize the single-objective optimization paradigm by introducing different objective functions for different players. Differentiab… (voir plus)le games often proceed by simultaneous or alternating gradient updates. In machine learning, games are gaining new importance through formulations like generative adversarial networks (GANs) and actor-critic systems. However, compared to single-objective optimization, game dynamics are more complex and less understood. In this paper, we analyze gradient-based methods with momentum on simple games. We prove that alternating updates are more stable than simultaneous updates. Next, we show both theoretically and empirically that alternating gradient updates with a negative momentum term achieves convergence in a difficult toy adversarial problem, but also on the notoriously difficult to train saturating GANs.
2019-04-11
Proceedings of the Twenty-Second International Conference on Artificial Intelligence and Statistics (publié)
proceedings.mlr.press
arxiv.org
Negative Momentum for Improved Game Dynamics
Gauthier Gidel
Reyhane Askari Hemmat
Mohammad Pezeshki
Gabriel Huang
Rémi LE PRIOL
Simon Lacoste-Julien
Ioannis Mitliagkas
Games generalize the single-objective optimization paradigm by introducing different objective functions for different players. Differentiab… (voir plus)le games often proceed by simultaneous or alternating gradient updates. In machine learning, games are gaining new importance through formulations like generative adversarial networks (GANs) and actor-critic systems. However, compared to single-objective optimization, game dynamics are more complex and less understood. In this paper, we analyze gradient-based methods with momentum on simple games. We prove that alternating updates are more stable than simultaneous updates. Next, we show both theoretically and empirically that alternating gradient updates with a negative momentum term achieves convergence in a difficult toy adversarial problem, but also on the notoriously difficult to train saturating GANs.
2018-07-12
ArXiv (preprint)
arxiv.org
arxiv.org
Frank-Wolfe Splitting via Augmented Lagrangian Method
Gauthier Gidel
Fabian Pedregosa
Simon Lacoste-Julien
Minimizing a function over an intersection of convex sets is an important task in optimization that is often much more challenging than mini… (voir plus)mizing it over each individual constraint set. While traditional methods such as Frank-Wolfe (FW) or proximal gradient descent assume access to a linear or quadratic oracle on the intersection, splitting techniques take advantage of the structure of each sets, and only require access to the oracle on the individual constraints. In this work, we develop and analyze the Frank-Wolfe Augmented Lagrangian (FW-AL) algorithm, a method for minimizing a smooth function over convex compact sets related by a "linear consistency" constraint that only requires access to a linear minimization oracle over the individual constraints. It is based on the Augmented Lagrangian Method (ALM), also known as Method of Multipliers, but unlike most existing splitting methods, it only requires access to linear (instead of quadratic) minimization oracles. We use recent advances in the analysis of Frank-Wolfe and the alternating direction method of multipliers algorithms to prove a sublinear convergence rate for FW-AL over general convex compact sets and a linear convergence rate for polytopes.
2018-03-31
ArXiv (preprint)
arxiv.org
arxiv.org
Frank-Wolfe Splitting via Augmented Lagrangian Method
Gauthier Gidel
Fabian Pedregosa
Simon Lacoste-Julien
Minimizing a function over an intersection of convex sets is an important task in optimization that is often much more challenging than mini… (voir plus)mizing it over each individual constraint set. While traditional methods such as Frank-Wolfe (FW) or proximal gradient descent assume access to a linear or quadratic oracle on the intersection, splitting techniques take advantage of the structure of each sets, and only require access to the oracle on the individual constraints. In this work, we develop and analyze the Frank-Wolfe Augmented Lagrangian (FW-AL) algorithm, a method for minimizing a smooth function over convex compact sets related by a "linear consistency" constraint that only requires access to a linear minimization oracle over the individual constraints. It is based on the Augmented Lagrangian Method (ALM), also known as Method of Multipliers, but unlike most existing splitting methods, it only requires access to linear (instead of quadratic) minimization oracles. We use recent advances in the analysis of Frank-Wolfe and the alternating direction method of multipliers algorithms to prove a sublinear convergence rate for FW-AL over general convex compact sets and a linear convergence rate for polytopes.
2018-01-01
AISTATS (publié)
proceedings.mlr.press
arxiv.org