Empower curiosity-driven research. Follow your convictions. Think not just about how to advance technology, but how to direct its use for good.
These were among the key messages delivered by Geoffrey Hinton, University Professor Emeritus of computer science at the University of Toronto and winner of the 2024 Nobel Prize in Physics, during an Oct. 8 press conference held by the university to mark his historic award.
Widely regarded as “the godfather of AI,” Hinton was named a co-winner of the prize — alongside John J. Hopfield of Princeton University — for his work on Boltzmann machines and artificial neural networks, which laid the groundwork for advancements in AI and stimulated new research directions in physics.
U of T President Meric Gertler hailed Hinton for having “a profound impact on multiple fields and disciplines,” crediting “his leadership and exemplary mentorship of young scholars” with helping U of T become a global leader in AI and machine learning.
“I think one cannot overstate the impact of a win like this on the ability of Canada, Toronto and the University of Toronto to be able to welcome talented newcomers, great students and wonderful faculty from across the country and around the world because of the recognition that arises with Geoff’s win,” President Gertler said.
For his part, Hinton echoed his remarks from earlier in the day that he was “flabbergasted” to receive the prize and pleased that the Nobel committee recognized the advancements in artificial neural networks.
He also answered questions about his influences, legacy and how it feels to go from being an obscure researcher who toiled in a largely forsaken field to a Nobel Laureate — and his advice for researchers who hope to one day follow in his footsteps.
Here are five key themes that emerged from Hinton's news conference:
His legacy
“I’m hoping AI will lead to tremendous benefits, to tremendous increases in productivity and to a better life for everybody. I’m convinced that it will do that in health care.
“My worry is that it may also lead to bad things, and in particular, when we get things more intelligent than ourselves, no one really knows whether we’re going to be able to control them.
“We don’t know how to avoid [catastrophic AI scenarios] at present. That’s why we urgently need more research. So I’m advocating that our best young researchers, or many of them, should work on AI safety and governments should force large companies to provide the computational facilities they need to do that.”
A collaborative effort
“I think of the prize as a recognition of a large community of people who worked on artificial neural networks for many years.
“I’d particularly like to acknowledge my two main mentors: David Rumelhart, with whom I worked on the backpropagation algorithm … and my colleague Terry Sejnowsky, who I worked with a lot in the 1980s on Boltzmann machines and who taught me a lot about the brain.
“I’d also like to acknowledge my students. I was particularly fortunate to have many clever students, much cleverer than me, who actually made things work. They’ve gone on to do great things.
“I should also acknowledge Yoshua Bengio and Yann LeCun who were close colleagues and very instrumental in developing this whole field.”
Canada’s research strengths
“I think the main thing about Canada as a place to do research is there isn’t as much money as there is in the U.S., but it uses its money quite wisely.
“In particular, the main funding council for this type of research, called NSERC, uses money for basic curiosity-driven research, and all of these advances in neural networks came out of basic curiosity-driven research — not out of throwing money at applied problems, but out of letting scientists follow their curiosity to try and understand things. And Canada’s quite good at that.”
Many thought he was wasting his time
“It was a lot of fun doing the research, but it was slightly annoying that many people — in fact, most people in the field of AI — said that neural networks would never work.
“They were very confident these things were a waste of time and we would never be able to learn complicated things — for example, understanding natural language — using neural networks. And they were wrong.”
Believe in your ideas
“My message is this: if you believe in something, don’t give up on it until you understand why that belief is wrong.
“Often, you believe in things and you eventually figure out why that’s a wrong thing to believe in. But so long as you believe in something and you can’t see why that’s wrong — like, ‘the brain has to work somehow so we have to figure out how it learns the connection strengths to make it work’ — keep working on it and don’t let people tell you it’s nonsense if you can’t see why it’s nonsense.”
— Original story by Rahul Kalvapalle for U of T News