‘if you want to make a device do something intelligent, you’ve got two options: You can program it, or it can learn. And people certainly weren’t programmed, so we had to learn. This had to be the right way to go.’
Deep learning pioneer Geoffrey Hinton says he was convinced all along ‘there’s no other way the brain can work.’ Hinton’s comments were in an interview with Wired magazine.
Hinton’s approach to deep learning is an homage to biology and how the brain works.
Deep learning processes data in multiple layers, using a ‘neural network’ comprised of huge numbers of nodes, called ‘neurons.’ Using the same word as brain cells is deliberate.
He quickly points out he’s not trying to reconstruct the brain. Instead, he considers brain functions as his inspiration.
An AI pioneer explains the evolution of neural networks
WIRED | May 13, 2019 | by Nicholas Thompson