Portrait of Hugo Larochelle

Hugo Larochelle

Scientific Director, Leadership Team
Adjunct professor, Université de Montréal, Department of Computer Science and Operations Research
Adjunct professor, McGill University, School of Computer Science
Research Topics
Deep Learning

Biography

Hugo Larochelle is the Scientific Director of Mila, one of the world’s leading artificial intelligence research institutes. With a community of close to 2,000 researchers and professionals, Mila has established itself as a pillar of the Canadian AI ecosystem with a reach that extends far beyond national borders.

As a pioneering researcher and industry leader, he has a unique perspective on both large-scale corporate research laboratories and Canada’s world-class academic AI community He built his academic foundation alongside two "Godfathers" of artificial intelligence: Yoshua Bengio and Geoffrey Hinton.

Over the years, his research has contributed several conceptual breakthroughs found in modern AI systems. His work on Denoising Autoencoders (DAE) identified the reconstruction of clean data from corrupted versions as a scalable paradigm for learning meaningful representations from large quantities of unlabeled data. Through models such as the Neural Autoregressive Distribution Estimator (NADE) and the Masked Autoencoder for Distribution Estimation (MADE), he helped popularize the neural autoregressive modeling paradigm now omnipresent in generative AI. Furthermore, his work on Zero-Data Learning of New Tasks introduced the now-standard concept of zero-shot learning.

He successfully bridged the gap between academia and industry by co-founding the startup Whetlab, which was acquired by Twitter in 2015. After a role at Twitter Cortex, he was recruited to lead Google's AI research lab in Montreal (Google Brain), now integrated into Google DeepMind. He remains an Adjunct Professor at the Université de Montréal and McGill University, and is a Canada CIFAR AI Chair, mentoring the next generation of AI researchers.

Alongside his role as Scientific Director at Mila, he also serves as Scientific Lead at Adaption Labs and advises the startups Tiptree Systems and Prizmal.

A father of four, Hugo Larochelle and his wife, Angèle St-Pierre, have also made multiple donations to the Université de Montréal and Université de Sherbrooke, particularly in AI for environmental sustainability. He also founded the Techaide conference, mobilizing Montreal's tech community to raise funds for the charity Centraide in its mission to fight poverty and social exclusion.

Current Students

PhD - Université de Montréal
Principal supervisor :
Professional Master's - McGill University
Collaborating Alumni - Université de Montréal
Principal supervisor :
Independent visiting researcher - McGill University
Postdoctorate - Polytechnique Montréal
Principal supervisor :

Publications

Brain Tumor Segmentation with Deep Neural Networks
Hierarchical Memory Networks
Memory networks are neural networks with an explicit memory component that can be both read and written to by the network. The memory is oft… (see more)en addressed in a soft way using a softmax function, making end-to-end training with backpropagation possible. However, this is not computationally scalable for applications which require the network to read from extremely large memories. On the other hand, it is well known that hard attention mechanisms based on reinforcement learning are challenging to train successfully. In this paper, we explore a form of hierarchical memory network, which can be considered as a hybrid between hard and soft attention memory networks. The memory is organized in a hierarchical structure such that reading from it is done with less computation than soft attention over a flat memory, while also being easier to train than hard attention over a flat memory. Specifically, we propose to incorporate Maximum Inner Product Search (MIPS) in the training and inference procedures for our hierarchical memory network. We explore the use of various state-of-the art approximate MIPS techniques and report results on SimpleQuestions, a challenging large scale factoid question answering task.