How Montreal became a nexus for deep learning
Powerful concept helps get big-picture products to market
All eyes are on Montreal these days as a hub for deep learning. “Clearly it’s a place where everybody wants to be if we want to tap into that talent,” says Nagraj Kashyap, corporate vice- president of Microsoft Ventures in San Francisco.
Montreal’s pre- eminence as a deep- learning centre can largely be attributed to the efforts of Yoshua Bengio, considered one of the three “co- fathers” of deeplearning technology. Bengio not only engaged in cutting- edge research at the Université de Montréal long before deep learning was considered viable; his work has spawned an ecosystem that many say is unrivalled in the artificial intelligence (AI) world.
That ecosystem includes the Montreal Institute for Learning Algorithms ( MILA), which has been funded by government and private sector parties, including Google and Microsoft among other tech notables. There’s also the IVADO ( L’institut de valorisation des données, otherwise known as the Institute for Data Valorisation), established in September 2016 from MILA funding. IVADO is a multi- disciplinary centre that acts as a link between academic research and business needs of organizations of all sizes, including startups.
Last but not least is the launch of Element AI in October 2016, an artificial intelligence startup factory where Bengio is one of the co-founders.
“In simple terms, the MILA does academic research, the IVADO does industrial research, and Element AI builds applications in partnership with industry for commercialization of AI,” says Jean- François Gagne, Element AI co-founder and CEO.
One noteworthy achievement for Element AI is that it was the first company to receive funding from Microsoft Ventures. Gagne says its intent is to help build companies from AI research at the Université de Montréal, as well as McGill University — a community of 150 students, professors and supporting staff dedicated to deep learning, which many claim is the largest in the world.
“Our business is really about collaborating with large corporations to build companies, create spinoffs and joint ventures, and launch them,” Gagne says. “We’re a bit different than typical incubators where people with technical skills come for support. We pro- vide both the technical skills and the technology so projects can get to market.”
Its efforts will gain significant traction in time, Kashyap says. “Element AI is connecting all businesses wanting to get AI talent. As more companies come and more get sold it will create a bigger ecosystem. In a few years from now, if Montreal is successful in creating new companies, ( that activity) will recirculate and create an even bigger hub for AI and deep learning.”
Gagne contends that the Montreal ecosystem has some unique elements. “What’s great about the MILA, for example, is that it has very few strings attached. We never give anyone exclusivity to the IT or anything else that might get in the way of moving and innovating faster. It’s totally open, which is really rare.”
The ecosystem has now achieved a critical mass that has attracted a lot of international players, he adds. For example Google has transformed its office into a third global AI lab; the other two are in London and California. “Microsoft didn’t have a (dedicated) research or development office in Montreal before investing in us, and acquiring Maluuba.” ( Maluuba is a Montreal- based deep learning research lab startup acquired by Microsoft in January 2017.)
Montreal has reached a crucial tipping point, Gagne believes. “We have a team of researchers on a permanent basis and a dynamic startup community. Last year, Montreal received the most amount of VC funding for technology. We surpassed Toronto and will again this year. One VC firm, Real Ventures, is now focusing on AI funding exclusively.’
International VCs are also frequent visitors. Gagne says they often play host to VCs from Asia, Europe and the U. S. “They are coming here on a regular basis. The amount of funding available is unprecedented.”
Many of the projects they are currently engaged in were generated from incubators and accelerators from across Canada and around the world, and have far exceeded initial expectations, he says. “We were expecting to be doing four projects by now. We’re already up to 10. Our employee numbers are already double what we projected and will be over 100 before the end of the year. The money we were thinking of raising over the next year will be (reached) next week.”
The most promising thing for Gagne is that Montreal’s success is helping keep startups closer to home. “We used to hear a lot about startups moving to the Silicon Valley to find the funding and the talent. It’s no longer a story I’m hearing. We’re actually seeing some coming back from California. It means the value is now staying here and will amplify itself over the next five years. It will be a different ecosystem than what we are seeing right now.”
LAST YEAR, MONTREAL RECEIVED (THE MOST) VC FUNDING FOR TECHNOLOGY.