Propulsé parle possible
Nous repoussons les limites de la connaissance afin de redéfinir ce qui est possible dans le domaine de l’intelligence artificielle.
-
Recherche fondamentale
Poursuivre de nouvelles pistes dans des domaines clés de la recherche sur l’IA, dont la vision par ordinateur, le traitement du langage naturel et l’apprentissage automatique
-
Recherche appliquée
Explorer comment l’IA peut être mise en œuvre dans des situations réelles, tester des prototypes et mettre au point des applications dans des domaines clés comme la robotique et l’interaction humain-ordinateur
-
Impact de l’IA
Étudier l’incidence de l’IA sur le monde, y compris l’éthique et l’explicabilité, ainsi que la possibilité d’utiliser l’IA pour relever des défis dans des domaines comme le changement climatique et les droits de la personne
Articles récemment publiés
-
Beyond Trivial Counterfactual Explanations with Diverse Valuable Explanations
Pau Rodriguez, Massimo Caccia, Alexandre Lacoste, Lee Zamparo, Issam Laradji, Laurent Charlin, David Vazquez
-
Disentanglement via Mechanism Sparsity Regularization: A New Principle for Nonlinear ICA
Sébastien Lachapelle, Pau Rodríguez López, Yash Sharma, Katie Everett, Rémi Le Priol, Alexandre Lacoste, Simon Lacoste-Julien
-
Variational Causal Networks: Approximate Bayesian Inference over Causal Structures
Yashas Annadani, Jonas Rothfuss, Alexandre Lacoste, Nino Scherrer, Anirudh Goyal, Yoshua Bengio, Stefan Bauer
-
OverNet: Lightweight Multi-Scale Super-Resolution With Overscaling Network
Parichehr Behjati, Pau Rodriguez, Armin Mehri, Isabelle Hupont, Carles Fernandez Tena, Jordi Gonzalez
-
Can Active Learning Preemptively Mitigate Fairness Issues?
Frédéric Branchaud-Charron, Parmida Atighehchian, Pau Rodríguez, Grace Abuhamad, Alexandre Lacoste
-
A Weakly Supervised Consistency-Based Learning Method for COVID-19 Segmentation in CT Images
Issam Laradji, Pau Rodriguez, Oscar Manas, Keegan Lensink, Marco Law, Lironne Kurzman, William Parker, David Vazquez, Derek Nowrouzezahrai
-
Stochastic Polyak Step-size for SGD: An Adaptive Learning Rate for Fast Convergence
Nicolas Loizou, Sharan Vaswani, Issam Hadj Laradji, Simon Lacoste-Julien
-
3D Perception With Slanted Stixels on GPU
Daniel Hernandez-Juarez, Antonio Espinosa, David Vazquez, Antonio M. Lopez, Juan C. Moure
-
Weakly Supervised Underwater Fish Segmentation Using Affinity LCFCN
Alzayat Saleh, Issam Laradji, Pau Rodriguez, Derek Nowrouzezahrai, Mostafa Rahimi Azghadi, David Vazquez
-
Seasonal Contrast: Unsupervised Pre-Training from Uncurated Remote Sensing Data
Oscar Mañas, Alexandre Lacoste, Xavier Giro-i-Nieto, David Vazquez, Pau Rodriguez
-
Touch-based Curiosity for Sparse-Reward Tasks
Sai Rajeswar, Cyril Ibrahim, Nitin Surya, Florian Golemo, David Vazquez, Aaron Courville, Pedro O. Pinheiro
-
A Deep Learning Localization Method for Measuring Abdominal Muscle Dimensions in Ultrasound Images
Alzayat Saleh, Issam H. Laradji, Corey Lammie, David Vazquez, Carol A Flavell, Mostafa Rahimi Azghadi
-
PICARD: Parsing Incrementally for Constrained Auto-Regressive Decoding from Language Models
Torsten Scholak, Nathan Schucher, Dzmitry Bahdanau
-
Synbols: Probing Learning Algorithms with Synthetic Datasets
Alexandre Lacoste, Pau Rodríguez, Frédéric Branchaud-Charron, Parmida Atighehchian, Massimo Caccia, Issam Laradji, Alexandre Drouin, Matt Craddock, Laurent Charlin, David Vázquez
-
DuoRAT: Towards Simpler Text-to-SQL Models
Torsten Scholak, Raymond Li, Dzmitry Bahdanau, Harm de Vries, Chris Pal
-
Differentiable Causal Discovery from Interventional Data
Philippe Brouillard, Sébastien Lachapelle, Alexandre Lacoste, Simon Lacoste-Julien, Alexandre Drouin
-
CLOSURE: Assessing Systematic Generalization of CLEVR models
Aaron Courville, Yoshua Bengio, Philippe Beaudoin, Shikhar Murty, Harm de Vries,Dzmitry Bahdanau
-
On Extractive and Abstractive Neural Document Summarization with Transformer Language Models
Sandeep Subramanian, Raymond Li, Jonathan Pilault, Christopher Pal
-
On the role of data in PAC-Bayes bounds
Gintare Karolina Dziugaite, Jonathan Frankle, Michael Carbin, Daniel M. Roy
-
Learning Data Augmentation with Online Bilevel Optimization for Image Classification
Saypraseuth Mounsaveng, Issam H. Laradji, Ismail Ben Ayed, David Vazquez, Marco Pedersoli
-
Adaptive Gradient Methods Converge Faster with Over-Parameterization (and you can do a line-search)
Sharan Vaswani, Issam H. Laradji, Frederik Kunstner,Si Yi Meng, Mark Schmidt, Simon Lacoste-Julien
-
DuoRAT: Towards Simpler Text-to-SQL Models
Torsten Scholak, Raymond Li, Dzmitry Bahdanau, Harm de Vries, Chrisopher Pal
-
Pruning Neural Networks at Initialization: Why are We Missing the Mark?
Jonathan Frankle, Gintare Karolina Dziugaite, Daniel M. Roy, Michael Carbin
-
Towards Ecologically Valid Research on Language User Interfaces.
Harm de Vries, Dzmitry Bahdanau, Christopher Manning
-
Differentiable Causal Discovery from Interventional Data
Philippe Brouillard, Sébastien Lachapelle, Alexandre Lacoste, Simon Lacoste-Julien, Alexandre Drouin
-
In search of robust measures of generalization
Gintare Karolina Dziugaite, Alexandre Drouin, Brayden (Brady) Neal, Nitarshan Rajkumar, Ethan Victor Caballero, Ioannis Mitliagkas, Daniel M. Roy
-
Synbols: Probing Learning Algorithms with Synthetic Datasets
Alexandre Lacoste, Pau Rodriguez, Frederic Branchaud, Parmida Atighehchian, Massimo Caccia, Alexandre Drouin, Issam H. Laradji, Matt Craddock, Laurent Charlin, David Vazquez
-
Online Fast Adaptation and Knowledge Accumulation: a New Approach to Continual Learning
Massimo Caccia, Pau Rodriguez, Oleksiy Ostapenko, Fabrice Normandin, Min Lin, Lucas Caccia, Issam H. Laradji, Irina Rish, David Vazquez, Laurent Charlin
-
Unsupervised Learning of Dense Visual Representations
Pedro O. Pinheiro, Amjad Almahairi, Ryan Y. Benmaleck, Florian Golemo, Aaron Courville
-
Measuring Systematic Generalization in Neural Proof Generation with Transformers
Nicolas Gontier, Koustuv Sinha, Siva Reddy, Christopher Pal
-
Promoting Coordination through Policy Regularization in Multi-Agent Deep Reinforcement Learning
Julien Roy, Paul Barde, Félix G. Harvey, Derek Nowrouzezahrai, Christopher Pal
-
Adversarial Soft Advantage Fitting: Imitation Learning without Policy Optimization
Paul Barde, Julien Roy, Wonseok Jeon, Joelle Pineau, Christopher Pal, Derek Nowrouzezahrai
-
Sharpened Generalization Bounds based on Conditional Mutual Information and an Application to Noisy-Gradient Iterative Algorithms
Gintare Karolina Dziugaite, Mahdi Haghifam, Jeffrey Negrea, Daniel M. Roy
-
An empirical study of loss landscape geometry and evolution of the data-dependent Neural Tangent Kernel
Gintare Karolina Dziugaite, Stanislav Fort, Daniel M. Roy, Surya Ganguli, Mansheej Paul, Sepideh Kharaghani
-
Gradient-Based Neural DAG Learning
Philippe Brouillard, Alexandre Drouin, Sébastien Lachapelle, Alexandre Lacoste, Simon Lacoste-Julien
-
The Variational Bandwidth Bottleneck: Stochastic Evaluation on an Information Budget
Anirudh Goyal, Yoshua Bengio, Matthew Botvinick, Sergey Levine -
Reinforcement Learning with Competitive Ensembles of Information-Constrained Primitives
Anirudh Goyal, Shagun Sodhani, Jonathan Binas, Xue Bin Peng, Sergey Levine, Yoshua Bengio -
A Meta-Transfer Objective for Learning to Disentangle Causal Mechanisms
Yoshua Bengio, Tristan Deleu, Nasim Rahaman, Nan Rosemary Ke, Sebastien Lachapelle, Olexa Bilaniuk, Anirudh Goyal, Christopher Pal -
Knowledge Hypergraphs: Extending Knowledge Graphs Beyond Binary Relations
Bahare Fatemi, Perouz Taslakian, David Vázquez, and David Poole
-
Fast and Furious Convergence: Stochastic Second Order Methods under Interpolation
Si Yi Meng, Sharan Vaswani, Issam Laradji, Mark Schmidt, and Simon Lacoste-Julien
-
Reinforcement Active Leanring for Image Segmentation
Arantxa Casanova, Pedro O. Pinheiro, Negar Rostamzadeh, and Christopher J. Pal
-
A Closer Look at the Optimization Landscapes of Generative Adversarial Networks
Hugo Berard, Gauthier Gidel, Amjad Almahairi, Pascal Vincent, and Simon Lacoste-Julien
-
N-BEATS: Neural basis expansion analysis for interpretable time series forecasting
Boris N. Oreshkin, Dmitri Carpov, Nicolas Chapados, and Yoshua Bengio
-
Slanted Stixels: A Way to Represent Steep Streets
Daniel Hernandez-Juarez, Lukas Schneider, Pau Cebrian, Antonio Espinosa, David Vazquez, AntonioM.López, Uwe Franke, Marc Pollefeys, and Juan C. Moure
-
Extending the spatial scale of land use regression models for ambient ultrafine particles using satellite images and deep convolutional neural networks
O.Pinheiro, Laura Minet, Marianne Hatzopoulou, and Scott Weichenthal
-
Pay attention to the activations: a modular attention mechanism for fine-grained image recognition
Pau Rodriguez, Diego Velazquez, Guillem Cucurull, Josep M. Gonfaus, F. Xavier Roca, and Jordi Gonzalez
-
In Defense of Uniform Convergence: Generalization via derandomization with an application to interpolating predictors
Jeffrey Negrea, Gintare Karolina Dziugaite,and Daniel M. Roy
-
How to Evaluate Machine Learning Approaches for Combinatorial Optimization: Application to the Travelling Salesman Problem
Antoine Francois, Quentin Cappart, and Louis-Martin Rousseau
-
Instance Segmentation with Point Supervision
Issam H. Laradji, Negar Rostamzadeh, Pedro O. Pinheiro, David Vazquez, and Mark Schmidt
-
Adversarial Computation of Optimal Transport Maps
Jacob Leygonie, Jennifer She, Amjad Almahairi, Sai Rajeswar, and Aaron Courville
-
Adaptive Deep Kernel Learning
Prudencio Tossou, Basile Dura, Mario Marchand, François Laviolette, and Alexandre Lacoste
-
Tackling Climate Change with Machine Learning
David Rolnick, Priya L. Donti, Lynn H. Kaack, Kelly Kochanski, Alexandre Lacoste, Kris Sankaran, Andrew Slavin Ross, Nikola Milojevic-Dupont, Natasha Jaques, Anna Waldman-Brown, Alexandra Luccioni, Tegan Maharaj, Evan D. Sherwin, S. Karthik Mukkavilli, Konrad P. Kording, Carla Gomes, Andrew Y. Ng, Demis Hassabis, John C. Platt, Felix Creutzig, Jennifer Chayes, and Yoshua Bengio
-
Finding and Visualizing Weaknesses of Deep Reinforcement Learning Agents
Christian Rupprecht, Cyril Ibrahim, and Christopher J. Pal
-
Structure Learning for Neural Module Networks
Vardaan Pahuja, Jie Fu, Sarath Chandar, and Christopher J. Pal
-
Stabilizing the Lottery Ticket Hypothesis
Jonathan Frankle, Karolina Dziugaite, Daniel M. Roy, and Michael Carbin
-
Manifold Preserving Adversarial Learning
Ousmane Amadou Dia, Elnaz Barshan, and Reza Babanezhad
-
Adversarial Mixup Resynthesis
Christopher Beckham, Sina Honari, Alex Lamb, Vikas Verma, Farnoosh Ghadiri, R Devon Hjelm, Yoshua Bengio, and Chris Pal
-
Visual Imitation with Reinforcement Learning using Recurrent Siamese Networks
Glen Berseth, and Christopher J. Pal
-
Reducing Noise in GAN Training with Variance Reduced Extragradient
Tatjana Chavdarova, Gauthier Gidel, François Fleuret, and Simon Lacoste-Julien
-
Painless Stochastic Gradient: Interpolation, Line-Search, and Convergence Rates
Sharan Vaswani Mila, Aaron Mishkin, Issam Laradji, Mark Schmidt, Gauthier Gidel, and Simon Lacoste-Julien
-
The Impact of Preprocessing on Arabic-English Statistical and Neural Machine Translation
Mai Oudah, Amjad Almahairi, and Nizar Habash
-
Active Domain Randomization
Bhairav Mehta, Manfred Diaz, Florian Golemo, Christopher J. Pal, and Liam Paull
-
AMP: Adaptive Masked Proxies for Few-Shot Segmentation
Mennatullah Siam, Boris Oreshkin, and Martin Jagersand
-
Domain-Adaptive Single-view 3D Reconstruction
Pedro O. Pinheiro, Negar Rostamzadeh, and Sungjin Ahn
-
Physical Adversarial Textures that Fool Visual Object Tracking
Rey Reza Wiyatno, and Anqi Xu
-
Where are the Masks: Instance Segmentation with Image-level Supervision
Issam H. Laradji, David Vazquez, and Mark Schmidt
-
Efficient Deep Gaussian Process Models for Variable-Sized Inputs
Issam H. Laradji, Mark Schmidt, Vladimir Pavlovic, and Minyoung Kim
-
On the impressive performance of randomly weighted encoders in summarization tasks
Jonathan Pilault, Jaehong Park, and Christopher Pal
-
Context-Aware Visual Compatibility Prediction
Guillem Cucurull, Perouz Taslakian, David Vazquez
-
Hierarchical Importance Weighted Autoencoders
Chin-Wei Huang, Kris Sankaran, Eeshan Dhekane, Alexandre Lacoste, and Aaron Courville
-
On Difficulties of Probability Distillation
Chin-Wei Huang, Faruk Ahmed, Kundan Kumar, Alexandre Lacoste, and Aaron Courville
-
Quaternion Recurrent Neural Networks
Titouan Parcollet, Mirco Ravanelli, Mohamed Morchid, Georges Linarès, Chiheb Trabelsi, Renato De Mori, and Yoshua Bengio
-
Systematic Generalization: What is Required and Can it Be Learned?
Dzmitry Bahdanau, Shikhar Murty, Michael Noukhovitch,Thien Huu Nguyen, Harm de Vries, and Aaron Courville
-
BabyAI: A Platform to Study the Sample Efficiency of Grounded Language Learning
Maxime Chevalier-Boisvert, Dzmitry Bahdanau, Salem Lahlou, Lucas Willems, Chitwan Saharia, Thien Huu Nguyen, and Yoshua Bengio
-
Probabilistic Planning with Sequential Monte Carlo methods
Alexandre Piche,Valentin Thomas,Cyril Ibrahim,Yoshua Bengio, and Chris Pal
-
On extractive and abstractive Neural Document Summarization with Transformer Language Models
Sandeep Subramanian, Raymond Li, Jonathan Pilault, and Christopher Pal
-
Improving Optimization Bounds Using Machine Learning: Decision Diagrams Meet Deep Reinforcement Learning
Quentin Cappart, Emmanuel Goutierre, David Bergman, and Louis-Martin Rousseau
Nous réunissons le meilleur du monde universitaire et de l’industrie pour faire avancer les connaissances de pointe.
-
Audacieux et percutant
Plus de 40 titulaires d'un PhD travaillent sur les plus grands défis de l’IA
-
Ouvert et collaboratif
Accélérer les découvertes signifie travailler de manière transversale et collaborative avec un réseau international de chercheurs universitaires et d’experts en IA
-
Expertise approfondie
Nos chercheurs proviennent d’un large éventail de disciplines et ont la possibilité de collaborer étroitement avec la communauté universitaire, ce qui leur permet d’accéder aux idées les plus récentes et aux meilleurs talents
-
Bienveillant
Étudier l’incidence de l’IA et élaborer des modèles non biaisés, équitables, précis et sûrs
Recherche en vedette
En collaboration avec Yoshua Bengio, nous avons mis au point N-Beats, un modèle de prévision de série chronologique doté d’une architecture neurale profonde qui a surpassé les gagnants de récentes compétitions et les références de l’industrie avec toute une série d’ensembles de données, notamment M4, M3 et TOURISM.
En plus d’offrir un plus grand degré d’exactitude, N-Beats n’a pas besoin d’être réglé avec précision. Il est rapide à entraîner et à déployer, et facile à interpréter.
Plonger dans l’inconnu
Nos équipes de recherche fondamentale se penchent sur des problèmes dont les solutions sont inconnues, dans des domaines où la recherche est en cours et où nous voulons faire avancer les connaissances de pointe.
-
Vision par ordinateur
Vision pour les agents incarnés, reconnaissance de nouvelles tâches et exploitation des connaissances acquises lors de tâches précédentes, apprentissage à partir d’une supervision minimale, apprentissage de représentations partagées
-
Traitement du langage naturel
Compréhension de documents, interface en langue naturelle, représentation et généralisation de textes, généralisation systématique, ancrage linguistique
-
Apprentissage machine
Théorie de l’apprentissage machine, réutilisabilité, robustesse et explicabilité, évolutivité