Portrait of Yoshua Bengio

Yoshua Bengio

Core Academic Member
Canada CIFAR AI Chair
Full Professor, Université de Montréal, Department of Computer Science and Operations Research Department
Founder and Scientific Advisor, Leadership Team
Research Topics
Causality
Computational Neuroscience
Deep Learning
Generative Models
Graph Neural Networks
Machine Learning Theory
Medical Machine Learning
Molecular Modeling
Natural Language Processing
Probabilistic Models
Reasoning
Recurrent Neural Networks
Reinforcement Learning
Representation Learning
Website
Google Scholar

Biography

*For media requests, please write to medias@mila.quebec.

For more information please contact Marie-Josée Beauchamp, Administrative Assistant at marie-josee.beauchamp@mila.quebec.

Yoshua Bengio is recognized worldwide as a leading expert in AI. He is most known for his pioneering work in deep learning, which earned him the 2018 A.M. Turing Award, “the Nobel Prize of computing,” with Geoffrey Hinton and Yann LeCun.

Bengio is a full professor at Université de Montréal, and the founder and scientific advisor of Mila – Quebec Artificial Intelligence Institute. He is also a senior fellow at CIFAR and co-directs its Learning in Machines & Brains program, serves as special advisor and founding scientific director of IVADO, and holds a Canada CIFAR AI Chair.

In 2019, Bengio was awarded the prestigious Killam Prize and in 2022, he was the most cited computer scientist in the world by h-index. He is a Fellow of the Royal Society of London, Fellow of the Royal Society of Canada, Knight of the Legion of Honor of France and Officer of the Order of Canada. In 2023, he was appointed to the UN’s Scientific Advisory Board for Independent Advice on Breakthroughs in Science and Technology.

Concerned about the social impact of AI, Bengio helped draft the Montréal Declaration for the Responsible Development of Artificial Intelligence and continues to raise awareness about the importance of mitigating the potentially catastrophic risks associated with future AI systems.

Current Students

Jamal Abou Haibeh
Collaborating Alumni - McGill University
Github
Google Scholar
Mohammed Abukalam
Collaborating Alumni - Université de Montréal
Github
Berkes Anaïs
Collaborating researcher - Cambridge University
Principal supervisor :
David Rolnick
Website
Github
Google Scholar
Rim Assouel
PhD - Université de Montréal
Stefan Bauer
Independent visiting researcher
Co-supervisor :
Guillaume Lajoie
Google Scholar
Paul Bertin
PhD - Université de Montréal
Joyce Chai
Independent visiting researcher
Principal supervisor :
Siva Reddy
Shahana Chatterjee
Collaborating researcher - N/A
Principal supervisor :
David Rolnick
Google Scholar
Xiaoyin Chen
PhD - Université de Montréal
Sanghyeok Choi
Collaborating researcher - KAIST
Website
Github
Google Scholar
Tristan Deleu
Collaborating Alumni - Université de Montréal
Website
Github
Google Scholar
Aniket Didolkar
PhD - Université de Montréal
Website
Github
Google Scholar
Abdessamad EL KABID
Collaborating Alumni - Université de Montréal
Co-supervisor :
Loubna Benabbou
Desmond Elliott
Independent visiting researcher
Principal supervisor :
Siva Reddy
Website
Google Scholar
Eric Elmoznino
PhD - Université de Montréal
Co-supervisor :
Guillaume Lajoie
Website
Github
Google Scholar
Léna Ezzine
PhD - Université de Montréal
Jean-Pierre Falet
PhD - Université de Montréal
Website
Github
Google Scholar
Leo Feng
PhD - Université de Montréal
Website
Google Scholar
Pietro Greiner
PhD
Google Scholar
Mohsin Hasan
PhD - Université de Montréal
Website
Github
Google Scholar
Edward Hu
PhD - Université de Montréal
Moksh Jain
PhD - Université de Montréal
Website
Github
Google Scholar
Thomas Jiralerspong
PhD - Université de Montréal
Principal supervisor :
Guillaume Lajoie
Website
Github
Google Scholar
Younesse Kaddar
Collaborating Alumni - Université de Montréal
Website
Github
Hyeonah Kim
Postdoctorate - Université de Montréal
Principal supervisor :
Alex Hernandez-Garcia
Salem Lahlou
Collaborating Alumni - Université de Montréal
Tabitha Edith Lee
Postdoctorate - Université de Montréal
Principal supervisor :
Glen Berseth
Website
Github
Google Scholar
Chenghao Liu
Collaborating Alumni
Website
Github
Google Scholar
Zhen Liu
Collaborating Alumni - Université de Montréal
Principal supervisor :
Liam Paull
Kanika Madan
PhD - Université de Montréal
Nikolay Malkin
Collaborating Alumni - Université de Montréal
Website
Github
Google Scholar
Cristian Dragos Manta
PhD - Université de Montréal
Co-supervisor :
Dhanya Sridhar
Github
Sarthak Mittal
PhD - Université de Montréal
Principal supervisor :
Guillaume Lajoie
Website
Github
Google Scholar
Jama Mohamud
PhD - Université de Montréal
Principal supervisor :
Mirco Ravanelli
Website
Github
Google Scholar
Cyrus Neary
Postdoctorate - Université de Montréal
Principal supervisor :
Glen Berseth
Website
Github
Google Scholar
Phong Nguyen
Independent visiting researcher - Université de Montréal
Padideh Nouri
PhD - Université de Montréal
Principal supervisor :
Doina Precup
Website
Github
Google Scholar
Ali Parviz
Collaborating researcher - Ying Wu Coll of Computing
Google Scholar
Lena Podina
Collaborating researcher - University of Waterloo
Principal supervisor :
David Rolnick
Google Scholar
Nassim Rahaman
Collaborating Alumni - Max-Planck-Institute for Intelligent Systems
Amine RAZIG
Collaborating researcher - Université de Montréal
Co-supervisor :
Loubna Benabbou
Website
Jarrid Rector-Brooks
PhD - Université de Montréal
Danyal REHMAN
Postdoctorate - Université de Montréal
Github
James Requeima
Independent visiting researcher - Université de Montréal
Oli RICHARDSON
Postdoctorate - Université de Montréal
Website
Google Scholar
Camille Rochefort-Boulanger
PhD - Université de Montréal
Principal supervisor :
Julie Hussin
Github
Abhik Roychoudhury Roychoudhury
Independent visiting researcher
Principal supervisor :
Siva Reddy
Github
Luca Scimeca
Postdoctorate - Université de Montréal
Website
Github
Google Scholar
Dragos Secrieru
Collaborating Alumni - Université de Montréal
Marcin Sendera
Collaborating Alumni - Université de Montréal
Github
Google Scholar
Divya Sharma
Postdoctorate
Co-supervisor :
Alex Hernandez-Garcia
Website
Github
Marco Stock
Independent visiting researcher - Technical University of Munich
Github
Mélisande Astrid Crystal Teng
PhD - Université de Montréal
Co-supervisor :
Hugo Larochelle
Website
Github
Ivan Titov
Independent visiting researcher
Principal supervisor :
Siva Reddy
Website
Google Scholar
Alex Tong
Collaborating Alumni - Université de Montréal
Website
Github
Google Scholar
Donna Vakalis
Postdoctorate - Université de Montréal
Co-supervisor :
David Rolnick
Website
Github
Google Scholar
Siddarth Venkatraman
PhD - Université de Montréal
Principal supervisor :
Glen Berseth
Website
Github
Google Scholar
Omar G. Younis
Collaborating researcher
Website
Github
Google Scholar
Taeyoung YUN
Collaborating researcher - Université de Montréal
Github
Tianyu Zhang
PhD - Université de Montréal
Website
Github
Google Scholar
Nicole Zhang
PhD - McGill University
Principal supervisor :
Mathieu Blanchette
Website
Github
Google Scholar
Dinghuai Zhang
PhD - Université de Montréal
Principal supervisor :
Aaron Courville
Website
Harry Zhao
Collaborating Alumni - McGill University
Principal supervisor :
Doina Precup
Website
Github
Google Scholar

Publications

Amortizing intractable inference in diffusion models for vision, language, and control
Siddarth Venkatraman
Moksh J. Jain
Luca Scimeca
Minsu Kim
Marcin Sendera
Mohsin Hasan
Luke Rowe
Sarthak Mittal
Pablo Lemos
Emmanuel Bengio
Alexandre Adam
Jarrid Rector-Brooks
Yoshua Bengio
Glen Berseth
Nikolay Malkin
Diffusion models have emerged as effective distribution estimators in vision, language, and reinforcement learning, but their use as priors … (see more)in downstream tasks poses an intractable posterior inference problem. This paper studies amortized sampling of the posterior over data,
2024-05-31
ArXiv (preprint)
doi.org
arxiv.org
Amortizing intractable inference in diffusion models for vision, language, and control
Siddarth Venkatraman
Moksh J. Jain
Luca Scimeca
Minsu Kim
Marcin Sendera
Mohsin Hasan
Luke Rowe
Sarthak Mittal
Pablo Lemos
Emmanuel Bengio
Alexandre Adam
Jarrid Rector-Brooks
Yoshua Bengio
Glen Berseth
Nikolay Malkin
Diffusion models have emerged as effective distribution estimators in vision, language, and reinforcement learning, but their use as priors … (see more)in downstream tasks poses an intractable posterior inference problem. This paper studies amortized sampling of the posterior over data,
2024-05-31
ArXiv (preprint)
doi.org
arxiv.org
Learning diverse attacks on large language models for robust red-teaming and safety tuning
Seanie Lee
Minsu Kim
Lynn Cherif
David Dobre
Juho Lee
Sung Ju Hwang
Kenji Kawaguchi
Gauthier Gidel
Yoshua Bengio
Nikolay Malkin
Moksh J. Jain
Red-teaming, or identifying prompts that elicit harmful responses, is a critical step in ensuring the safe and responsible deployment of lar… (see more)ge language models (LLMs). Developing effective protection against many modes of attack prompts requires discovering diverse attacks. Automated red-teaming typically uses reinforcement learning to fine-tune an attacker language model to generate prompts that elicit undesirable responses from a target LLM, as measured, for example, by an auxiliary toxicity classifier. We show that even with explicit regularization to favor novelty and diversity, existing approaches suffer from mode collapse or fail to generate effective attacks. As a flexible and probabilistically principled alternative, we propose to use GFlowNet fine-tuning, followed by a secondary smoothing phase, to train the attacker model to generate diverse and effective attack prompts. We find that the attacks generated by our method are effective against a wide range of target LLMs, both with and without safety tuning, and transfer well between target LLMs. Finally, we demonstrate that models safety-tuned using a dataset of red-teaming prompts generated by our method are robust to attacks from other RL-based red-teaming approaches.
2024-05-28
ArXiv (preprint)
doi.org
arxiv.org
Learning diverse attacks on large language models for robust red-teaming and safety tuning
Seanie Lee
Minsu Kim
Lynn Cherif
David Dobre
Juho Lee
Sung Ju Hwang
Kenji Kawaguchi
Gauthier Gidel
Yoshua Bengio
Nikolay Malkin
Moksh J. Jain
Red-teaming, or identifying prompts that elicit harmful responses, is a critical step in ensuring the safe and responsible deployment of lar… (see more)ge language models (LLMs). Developing effective protection against many modes of attack prompts requires discovering diverse attacks. Automated red-teaming typically uses reinforcement learning to fine-tune an attacker language model to generate prompts that elicit undesirable responses from a target LLM, as measured, for example, by an auxiliary toxicity classifier. We show that even with explicit regularization to favor novelty and diversity, existing approaches suffer from mode collapse or fail to generate effective attacks. As a flexible and probabilistically principled alternative, we propose to use GFlowNet fine-tuning, followed by a secondary smoothing phase, to train the attacker model to generate diverse and effective attack prompts. We find that the attacks generated by our method are effective against a wide range of target LLMs, both with and without safety tuning, and transfer well between target LLMs. Finally, we demonstrate that models safety-tuned using a dataset of red-teaming prompts generated by our method are robust to attacks from other RL-based red-teaming approaches.
2024-05-28
ArXiv (preprint)
doi.org
arxiv.org
Learning diverse attacks on large language models for robust red-teaming and safety tuning
Seanie Lee
Minsu Kim
Lynn Cherif
David Dobre
Juho Lee
Sung Ju Hwang
Kenji Kawaguchi
Gauthier Gidel
Yoshua Bengio
Nikolay Malkin
Moksh J. Jain
Red-teaming, or identifying prompts that elicit harmful responses, is a critical step in ensuring the safe and responsible deployment of lar… (see more)ge language models (LLMs). Developing effective protection against many modes of attack prompts requires discovering diverse attacks. Automated red-teaming typically uses reinforcement learning to fine-tune an attacker language model to generate prompts that elicit undesirable responses from a target LLM, as measured, for example, by an auxiliary toxicity classifier. We show that even with explicit regularization to favor novelty and diversity, existing approaches suffer from mode collapse or fail to generate effective attacks. As a flexible and probabilistically principled alternative, we propose to use GFlowNet fine-tuning, followed by a secondary smoothing phase, to train the attacker model to generate diverse and effective attack prompts. We find that the attacks generated by our method are effective against a wide range of target LLMs, both with and without safety tuning, and transfer well between target LLMs. Finally, we demonstrate that models safety-tuned using a dataset of red-teaming prompts generated by our method are robust to attacks from other RL-based red-teaming approaches.
2024-05-28
ArXiv (preprint)
doi.org
arxiv.org
Learning diverse attacks on large language models for robust red-teaming and safety tuning
Seanie Lee
Minsu Kim
Lynn Cherif
David Dobre
Juho Lee
Sung Ju Hwang
Kenji Kawaguchi
Gauthier Gidel
Yoshua Bengio
Nikolay Malkin
Moksh J. Jain
Red-teaming, or identifying prompts that elicit harmful responses, is a critical step in ensuring the safe and responsible deployment of lar… (see more)ge language models (LLMs). Developing effective protection against many modes of attack prompts requires discovering diverse attacks. Automated red-teaming typically uses reinforcement learning to fine-tune an attacker language model to generate prompts that elicit undesirable responses from a target LLM, as measured, for example, by an auxiliary toxicity classifier. We show that even with explicit regularization to favor novelty and diversity, existing approaches suffer from mode collapse or fail to generate effective attacks. As a flexible and probabilistically principled alternative, we propose to use GFlowNet fine-tuning, followed by a secondary smoothing phase, to train the attacker model to generate diverse and effective attack prompts. We find that the attacks generated by our method are effective against a wide range of target LLMs, both with and without safety tuning, and transfer well between target LLMs. Finally, we demonstrate that models safety-tuned using a dataset of red-teaming prompts generated by our method are robust to attacks from other RL-based red-teaming approaches.
2024-05-28
ArXiv (preprint)
doi.org
arxiv.org
Estimating Expectations without Sampling: Neural Stein Estimation
Mohsin Hasan
Dinghuai Zhang
Cheikh Ahmed
Awa Khouna
Yoshua Bengio
We propose a method for estimating the expected value of a given function …
2024-05-27
approximateinference.org/AABI/2024/Symposium (accepted)
openreview.net
openreview.net
Attention as an RNN
Leo Feng
Frederick Tung
Hossein Hajimirsadeghi
Mohamed Osama Ahmed
Yoshua Bengio
Greg Mori
2024-05-22
ArXiv (preprint)
doi.org
arxiv.org
Attention as an RNN
Leo Feng
Frederick Tung
Hossein Hajimirsadeghi
Mohamed Osama Ahmed
Yoshua Bengio
Greg Mori
The advent of Transformers marked a significant breakthrough in sequence modelling, providing a highly performant architecture capable of le… (see more)veraging GPU parallelism. However, Transformers are computationally expensive at inference time, limiting their applications, particularly in low-resource settings (e.g., mobile and embedded devices). Addressing this, we (1) begin by showing that attention can be viewed as a special Recurrent Neural Network (RNN) with the ability to compute its \textit{many-to-one} RNN output efficiently. We then (2) show that popular attention-based models such as Transformers can be viewed as RNN variants. However, unlike traditional RNNs (e.g., LSTMs), these models cannot be updated efficiently with new tokens, an important property in sequence modelling. Tackling this, we (3) introduce a new efficient method of computing attention's \textit{many-to-many} RNN output based on the parallel prefix scan algorithm. Building on the new attention formulation, we (4) introduce \textbf{Aaren}, an attention-based module that can not only (i) be trained in parallel (like Transformers) but also (ii) be updated efficiently with new tokens, requiring only constant memory for inferences (like traditional RNNs). Empirically, we show Aarens achieve comparable performance to Transformers on
2024-05-22
ArXiv (preprint)
doi.org
arxiv.org
Attention as an RNN
Leo Feng
Frederick Tung
Hossein Hajimirsadeghi
Mohamed Osama Ahmed
Yoshua Bengio
Greg Mori
The advent of Transformers marked a significant breakthrough in sequence modelling, providing a highly performant architecture capable of le… (see more)veraging GPU parallelism. However, Transformers are computationally expensive at inference time, limiting their applications, particularly in low-resource settings (e.g., mobile and embedded devices). Addressing this, we (1) begin by showing that attention can be viewed as a special Recurrent Neural Network (RNN) with the ability to compute its \textit{many-to-one} RNN output efficiently. We then (2) show that popular attention-based models such as Transformers can be viewed as RNN variants. However, unlike traditional RNNs (e.g., LSTMs), these models cannot be updated efficiently with new tokens, an important property in sequence modelling. Tackling this, we (3) introduce a new efficient method of computing attention's \textit{many-to-many} RNN output based on the parallel prefix scan algorithm. Building on the new attention formulation, we (4) introduce \textbf{Aaren}, an attention-based module that can not only (i) be trained in parallel (like Transformers) but also (ii) be updated efficiently with new tokens, requiring only constant memory for inferences (like traditional RNNs). Empirically, we show Aarens achieve comparable performance to Transformers on
2024-05-22
ArXiv (preprint)
doi.org
arxiv.org
Attention as an RNN
Leo Feng
Frederick Tung
Hossein Hajimirsadeghi
Mohamed Osama Ahmed
Yoshua Bengio
Greg Mori
2024-05-22
ArXiv (preprint)
doi.org
arxiv.org
Attention as an RNN
Leo Feng
Frederick Tung
Hossein Hajimirsadeghi
Mohamed Osama Ahmed
Yoshua Bengio
Greg Mori
The advent of Transformers marked a significant breakthrough in sequence modelling, providing a highly performant architecture capable of le… (see more)veraging GPU parallelism. However, Transformers are computationally expensive at inference time, limiting their applications, particularly in low-resource settings (e.g., mobile and embedded devices). Addressing this, we (1) begin by showing that attention can be viewed as a special Recurrent Neural Network (RNN) with the ability to compute its \textit{many-to-one} RNN output efficiently. We then (2) show that popular attention-based models such as Transformers can be viewed as RNN variants. However, unlike traditional RNNs (e.g., LSTMs), these models cannot be updated efficiently with new tokens, an important property in sequence modelling. Tackling this, we (3) introduce a new efficient method of computing attention's \textit{many-to-many} RNN output based on the parallel prefix scan algorithm. Building on the new attention formulation, we (4) introduce \textbf{Aaren}, an attention-based module that can not only (i) be trained in parallel (like Transformers) but also (ii) be updated efficiently with new tokens, requiring only constant memory for inferences (like traditional RNNs). Empirically, we show Aarens achieve comparable performance to Transformers on
2024-05-22
ArXiv (preprint)
doi.org
arxiv.org