Learning in Machines and Brains
Project Overview
The CIFAR Learning in Machines & Brains program brings together 41 researchers to examine what consciousness is and how it comes about. The program is part of the Canadian Institute for Advanced Research, which was founded in 1982 to bring together researchers from across disciplines and borders to address important questions that affect humanity. Fellows in CIFAR programs come together regularly to collaborate in interdisciplinary groups which give rise to new insights. The program is co-directed by Yoshua Bengio of the Universite de Montreal and Yann LeCun of New York University. The program is helping to revolutionize the field of artificial intelligence and create computers that can think more like people – that can recognize faces, understand what is happening in a picture or video, and comprehend the meaning of language.
Research by members of the program has helped usher in a new generation of powerful AI, and members work closely with top technology firms including Google, Facebook and Microsoft. In one recent advance, Senior Fellow Max Welling of the University of Amsterdam and colleagues applied a deep learning technique called scattering networks to the analysis of Alzheimer’s disease from brain MRI data. The study showed that in a sem-supervised learning setting, it is possible to predict with 83 per cent certainty whether a patient diagnosed with mild cognitive impairment will develop Alzheimer’s disease. In another result, Co-Director Bengio made progress in showing how back-propagation – an error correction method used to train deep learning neural networks – could be linked to learning in the brain. His work uncovered new theorems and simulation results that bring us closer to a plausible explanation of how the brain could learn in a similar to way to backpropagation.
Principal Investigator
Yoshua Bengio , Université de Montréal
Team Members
Yann Lecun, Facebook AI Research
Francis Bach, Inria
Aaron Courville, Université de Montréal
Nando de Freitas, University of Oxford
James DiCarlo, Massachusetts Institute of Technology
David Fleet, University of Toronto
Brendan Frey, University of Toronto
Surya Ganguli, Stanford University
Zaid Harchaoui, Inria
Aapo Hyvarinen, University of Helsinki
Hugo Larochelle, Université de Sherbrooke
Honglak Lee, University of Michigan
Christopher Manning, Stanford University
Roland Memisevic, Université de Montréal
Andrew Ng, Stanford University
Bruno Olshausen, University of California, Berkeley
Ruslan Salakhutdinov, University of Toronto
Mark Schmidt, University of British Columbia
Eero Simoncelli, New York University
Josef Sivic, Inria
Ilya Sutskever, OpenAI
Richard Sutton, University of Alberta
Antonio Torralba, Massachusetts Institute of Technology
Pascal Vincent, Université de Montréal
Yair Weiss, The Hebrew University of Jerusalem
Max Welling, University of Amsterdam
Christopher Williams, The University of Edinburgh
Richard Zemel, University of Toronto
Partners and Donors
CIFAR