Deep Learning Summit
Montreal

“We need fundamental additions to this toolbox we have created to reach machines that operate at the level of true human understanding,” Dr. Bengio said. Though these systems have undeniably accelerated the progress of artificial intelligence, they are still a very long way from true intelligence. But Drs. Hinton, LeCun, and Bengio believe that new ideas will come.

Led by Hugo Larochelle and Matthew Jackson, from the Canadian Institute For Advanced Research, a team of international scientists and technologists are making the case for the creation of a new scientific discipline termed machine behaviour. This, they argue, is necessary to advance our understanding of how artificial intelligence will affect society, culture, the economy and politics.

CMU Associate Research Professor Simon Lucey uses himself as an example to show off his facial mapping software developed for an online eyeglass retailer. It uses a phone camera to take a picture of the users face, a cloud computer creates a 3-D map of the face and finally it applies the selected pair of glasses to the shopper's face.

Headed by Professor Simon Lacoste-Julien, Professor at the Université de Montréal and member of Mila, the SAIT AI (Samsung Advanced Institute of Technology Artificial Intelligence) Lab Montreal will conduct fundamental research on voice recognition, machine translation and natural language understanding (developing the capability of machines to understand the meaning of a text).

The new DeepMind initiative in Montreal will be led by Doina Precup, associate professor in the school of computer science at McGill, who will be bringing her expertise in reinforcement learning. Precup is also Sr fellow of the Canadian Institute for Advanced Research and a member of MILA.

Jonathan Frankle, the MIT PhD student who coauthored the paper "The Lottery Ticket Hypothesis" asks “If you need way fewer connections than what you started with,” he says, “why can’t we just train the smaller network without the extra connections?” Turns out you can.

RealityEngines.AI has been focused on dataset augmentation. Dataset augmentation can be particularly helpful, for instance, with banks that are trying to stop certain fraud cases but only have a few examples of past instances for AI models to learn from.