Neural networks are the cornerstone of machine learning and the basis for a lot of other artificial intelligence work.
It’s only really in the last few years that AI has hit the mainstream – reinforced by movies like Her and Channel 4’s TV series Humans but underpinning much of that (currently fictional) future is the idea that computer systems can learn.
The YouTube clip above shows cognitive psychologist and computer scientist Geoffrey Hinton, who splits his time between Google and the University of Toronto, explaining the aim of artificial neural networks nearly a decade ago.
I confess – the complexities of neural networks are perhaps a little beyond me, but it’s nonetheless amazing to vaguely grasp how Google’s approached making its products smarter, and in some cases you can see how the examples and theories discussed have led directly to changes in some of Google’s projects.
Make yourself a drink, sit back and get a little smarter this Sunday.