Speakers

Register
Vitória Barin Pacela

Vitória Barin Pacela

PhD student

Mila, Quebec AI Institute · DIRO, Université de Montréal · Meta (FAIR), Montreal

Bio Vitória is a Ph.D. student at Mila and the Université de Montréal, supervised by Professor Simon Lacoste-Julien. She is also a visiting researcher at Meta (FAIR) in Montreal, supervised by Professor Pascal Vincent. Vitória is broadly interested in causal representation learning.

Introduction to Probability


Website
Tristan Deleu

Tristan Deleu

PhD student

Mila, Quebec AI Institute · DIRO, Université de Montréal

Bio Tristan is a Ph.D. candidate at Mila & Université de Montréal, under the supervision of Yoshua Bengio. His research interests include Probabilistic Modeling, Structure Learning, Meta-Learning, Few-shot Learning, and Reinforcement Learning.

Probabilistic Modeling


Website
Sébastien Lachapelle

Sébastien Lachapelle

Research scientist

Samsung's SAIT AI Lab (SAIL) · Mila, Quebec AI Institute · DIRO, Université de Montréal

Bio Sébastien is a research scientist at Samsung's SAIT AI Lab (SAIL) in Montreal as well as a final year Ph.D. student supervised by Simon Lacoste-Julien. He has graduated from Université de Montréal with a Bachelor's degree in Mathematics and Economics. Sébastien's current focus is on questions of identifiability in representation learning. He also did work on causal structure learning with continuous-constrained optimization methods.

Probabilistic Inference


Website
Dr. Pablo Lemos

Pablo Lemos

Postdoctoral Research Fellow

Mila, Quebec AI Institute · CIELA institute, Université de Montréal

Bio Pablo is a postdoctoral Research Fellow in cosmology and machine learning at the Montrel Institute for Learning Algorithms (Mila) and the CIELA institute at the University of Montreal. His work focuses on applying machine learning tools to various problems in astrophysics and cosmology. He is very interested in graph neural networks, symbolic regression and simulation-based inference amongst other things.

Sampling Methods


Website
Padideh Nouri

Padideh Nouri

Mila, Quebec AI Institute

Bio

Sampling with Reinforcement Learning


Website
Prof. Yoshua Bengio

Yoshua Bengio

Scientific Director of Mila, Turing Award 2018

Mila, Quebec AI Institute · DIRO, Université de Montréal

Bio Recognized worldwide as one of the leading experts in artificial intelligence, Yoshua Bengio is most known for his pioneering work in deep learning, earning him the 2018 A.M. Turing Award, “the Nobel Prize of Computing,” with Geoffrey Hinton and Yann LeCun. He is a Full Professor at Université de Montréal, and the Founder and Scientific Director of Mila – Quebec AI Institute. He co-directs the CIFAR Learning in Machines & Brains program as Senior Fellow and acts as Scientific Director of IVADO. In 2019, he was awarded the prestigious Killam Prize and in 2022, became the computer scientist with the highest h-index in the world. He is a Fellow of both the Royal Society of London and Canada, Knight of the Legion of Honor of France and Officer of the Order of Canada. Concerned about the social impact of AI and the objective that AI benefits all, he actively contributed to the Montreal Declaration for the Responsible Development of Artificial Intelligence.

Introductory Remarks


Website
Dr. Emmanuel Bengio

Emmanuel Bengio

Senior Machine Learning Scientist

Recursion

Bio Emmanuel Bengio is a Sr ML Scientist at Valence Labs, working on the intersection of GFlowNets and de-novo drug design. He did his PhD under Joelle Pineau and Doina Precup at McGill/Mila, focusing on understanding generalization in deep RL.

Introduction to GFlowNets: Part 1


Website
Dr. Nikolay Malkin

Nikolay Malkin

Postdoctoral Fellow

Mila, Quebec AI Institute · DIRO, Université de Montréal

Bio Nikolay Malkin is a postdoctoral researcher at Mila – Québec AI Institute and Université de Montréal. His research interests include probabilistic inference algorithms for structured latent variables, induction of compositional structure in generative models, and applications to vision and language modeling. He received his PhD in mathematics from Yale University in 2021 and therefore views human-like symbolic and formal reasoning as a long-term aspiration for AI systems.

Introduction to GFlowNets: Part 2


Website
Jarrid Rector-Brooks

Jarrid Rector-Brooks

PhD Student

Mila, Quebec AI Institute · DIRO, Université de Montréal

Bio

Training GFlowNets


Website
Moksh Jain

Moksh Jain

PhD Student

Mila, Quebec AI Institute · DIRO, Université de Montréal

Bio Moksh Jain is a PhD student at Mila and Université de Montréal supervised by Yoshua Bengio. His research deals with questions around probabilistic inference and experimental design using tools from deep learning, with a focus on applications to accelerate scientific discovery.

Parameterization and Conditioning


Website
Léna Néhale Ezzine

Léna Néhale Ezzine

PhD Student

Mila, Quebec AI Institute · DIRO, Université de Montréal

Bio

Maximum Likelihood Training


Salem Lahlou

Salem Lahlou

PhD Student

Mila, Quebec AI Institute · DIRO, Université de Montréal

Bio Salem Lahlou is a last year PhD candidate at Mila (Université de Montréal) under the supervision of Yoshua Bengio. His research interests includes reinforcement learning, uncertainty estimation, and probabilistic modelling. Recently, he has been involved in the theory of generative flow networks (GFlowNets). Prior to his PhD, he studied applied mathematics in École Polytechnique and statistical learning in ENS Paris-Saclay, he did research in game theory and operations research in IBM Research Singapore, and worked as a data scientist in Booking.com in Amsterdam.

Continuous GFlowNets In this talk, we will delve into the mathematical tools required for generalizing GFlowNets to continuous, or mixed discrete-continuous, state spaces. We will see how the resulting algorithms apply to different settings, including stochastic control and Bayesian structure learning.


Website
Joseph Viviano

Joseph Viviano

ML Scientist

Mila, Quebec AI Institute

Bio Joseph Viviano is a ML Scientist at Mila, working on tool development for AI for science applications, and applications of GFlowNets in drug discovery. He holds degrees in Psychology, Biology, and Computer Science.

Hands-on: Training GFlowNets The hands-on sessions will focus on building understanding of the practical side of implementing GFlowNets.


Website
Alex Hernandez-Garcia

Alex Hernandez-Garcia

Postdoctoral scientist

Mila, Quebec AI Institute

Bio Alex Hernandez-Garcia is a postdoc at Mila, interested in the fundamental aspects of learning, both in brains and machines and in applications of machine learning to accelerate scientific discoveries to tackle the climate crisis.

Live coding of a GFlowNet environment Alex will code a GFlowNet environment from scratch, in the gflownet repository: github.com/alexhernandezgarcia/gflownet


Website Twitter
Alexandra Volokhova

Alexandra Volokhova

PhD student

Mila, Quebec AI Institute

Bio Alexandra is a third-year PhD student at Mila working on application of GFlowNet to drugs and materials discovery. She is interested in developing and applying fundamental machine learning for tackling socially important problems, such pandemics and climate change.

Molecular and protein conformation generation


Website
Mizu Nishikawa-Toomey

Mizu Nishikawa-Toomey

PhD student

Mila, Quebec AI Institute

Bio I am a 3rd year PhD student at Mila, supervised by Laurent Charlin and Dhanya Sridhar. I'm interested in active learning, uncertainty quantification and learning causal relations using machine learning.

Causal discovery We propose Variational Bayes GFlowNet for learning the distribution over causal structures and mechanism parameters.


Website
Dinghuai Zhang

Dinghuai Zhang

PhD student

Mila, Quebec AI Institute

Bio Dinghuai Zhang is a PhD candidate at Mila, advised by Prof. Aaron Courville and Prof. Yoshua Bengio. His research focuses on the intersection of probabilistic inference and scientific discovery.

Graph Combinatorial Problems


Website