Press "Enter" to skip to content


Post-doc Openings at Mila: Yoshua Bengio is leading a group of Mila professors working on machine learning for drug discovery (Jian Tang, Doina Precup, Pierre-Luc Bacon, Sarath Chandar, Amine Emad, Guy Wolf, Mathieu Blanchette, and myself) and looking for a couple of postdocs. One of them could play more of a project management role, coordinating projects and supervising junior students. Please fill the form here and write to Yoshua.

See this 2021 talk on ML for searching in the space of drugs and biological or chemical experiments.

Job Openings at Mila: I am looking for a project manager on the drug discovery projects.

My most exciting recent work is on what we call GFlowNets. It emerged from our desire to sample a set of good and diverse candidates for discovering new antiviral molecules, but it turns out to be a radically different way of thinking about probabilistic modeling, a very different kind of neural net tool for generating and operating on sets and graphs, a way to estimate what would otherwise be intractable marginalizations, a radically different way of performing credit assignment at the macro scale (i.e., different from backprop but closer to temporal-difference learning), and finally what I think may become a key ingredient for system-2 deep learning enabling causal discovery and reasoning. See the first paper on GFlowNets accepted for NeurIPS’2021 (see and the GFlowNet Foundations paper:

My current fundamental research vision was summarized by my Posner lecture at NeurIPS 2019:  From System 1 Deep Learning to System 2 Deep Learning, December 11th, 2019. Video with synchronized slides here.

Tutorial at IJCAI’2018 on Deep Learning for AI, July 13th, 2018.

Call for an International Ban on the Weaponization of Artificial Intelligence; AI Researchers ask the Canadian government to act at the UN in an open letter which you can sign too.

Video of my keynote at the first Cognitive and Computational Neuroscience conference at Columbia University on September 8, 2017.

Montreal AI Symposium

Montreal Deep Learning Summit, 10-11 Oct 2017, with Geoff Hinton, Yann LeCun and Yoshua Bengio

2017 Montreal Deep Learning Summer School and Reinforcement Learning Summer School and its video lectures 

Participation at the Beneficial AI Conference with video of my presentation.

Introductory articles in Scientific American about Deep Learning and AI.

More research highlights and selected recent papers

Nature paper on Deep Learning by Yann LeCun, Yoshua Bengio and Geoff Hinton (pdf)

NIPS’2015 Deep Learning Tutorial and the block of slides for the Vision part

NIPS’2014 Deep Learning and Representation Learning Workshop

Deep Learning – an MIT Press book now for sale here

ICLR: the International Conference on Learning Representations

2012 Review paper (published 2013 in PAMI) on Representation Learning

2012 Review paper on Practical recommendations for gradient-based training of deep architectures