Yoshua Bengio

Yoshua Bengio is a Canadian computer scientist, born in Paris, France, 5 March 1964. He grew up in France and in Quebec, Canada. He received his Bachelor degree in Engineering in 1986, Master of Science in Computer Science in 1988 and PhD in Computer Science in 1991 from McGill University, Montreal, Quebec, Canada. He was a post-doctoral fellow at M.I.T. Brain and Cognitive Sciences with Michael I. Jordan and AT&T Bell Labs with the group of Vladimir Vapnik, Larry Jackel, and Yann LeCun.

Bengio is known for his works on artificial neural networks and deep learning. He contributed a series of papers showing the fundamental limitations of gradient-based learning for parametrized dynamical systems such as recurrent neural networks and Hidden Markov Models (HMMs) and their combination in discriminant learning algorithms from his post-doctoral studies.

Bengio is a Professor in the Department of Computer Science and Operations Research at Université de Montréal, Canada. He is Canada Research Chair in Statistical Learning Algorithms. He is also head of the Montreal Institute for Learning Algorithms (MILA), Canadian Institute for Advanced Research (CIFAR) Program co-director of the program on Learning in Machines and Brains.

He is the action editor for the Journal of Machine Learning Research, editor for Foundations and Trends in Machine Learning, associate editor for the Machine Learning Journal and the IEEE Transactions on Neural Networks and associate editor for the Neural Computation journal.

Together with Yann LeCun, they created the International Conference on Learning Representations (ICLR). He organized and co-organized numerous other events, mainly the deep learning workshops and symposia at NIPS and ICML since 2007. He was Program Chair for NIPS 2008 and General Chair for NIPS 2009. NIPS is the flagship conference in the fields of learning algorithms and neural computation. He is Officer of the Order of Canada and a member of the Royal Society of Canada.

He has been pursuing to understand the principles of learning that will someday give humankind computers with intelligence. He focused on the major limitations of current learning algorithms to help develop algorithms avoiding these limitations. In the 1900's he outlined the difficulties of learning to represent circumstances. Recently, he focused on the limitations of shallow architectures and strategies for dealing with the hard optimization and inference challenges presented by deeper architectures.

His research work is widely cited over 80,000 citations found by Google Scholar in September 2017.