Geoffrey Hinton
Geoffrey Hinton spent forty years on an approach almost everyone thought was a dead end, and then it worked. Born in London in 1947 into a family of distinguished scientists — his great-great-grandfather was George Boole — he read experimental psychology at Cambridge after abandoning physiology and philosophy, and took a doctorate in artificial intelligence at Edinburgh in 1978 at a time when neural networks were regarded as discredited. The field had been dominated since the 1960s by symbolic AI, and Minsky and Papert's book on the limits of the perceptron had killed funding for the alternative. Hinton's conviction was that intelligence emerges from learned connection strengths in a network, not from hand-written rules. With David Rumelhart and Ronald Williams he published the backpropagation paper in 1986, which showed how to train networks with multiple layers by propagating error backwards. It is the algorithm behind essentially all modern deep learning. He moved to Toronto in 1987, partly to avoid military funding, and worked through two decades in which the approach remained marginal for want of data and computing power. Both arrived. In 2012 his students Alex Krizhevsky and Ilya Sutskever won the ImageNet competition with AlexNet by an enormous margin using GPUs, and the field changed within a year. He shared the Turing Award in 2018 and the Nobel Prize in Physics in 2024. He left Google in 2023 to speak freely about the risks, saying he partly regretted his life's work. He has argued since that systems more capable than their designers may not stay controllable, and that the timeline is shorter than he once assumed.
- Lived: 1947 CE – ?
- Nationality: British
- Roles: computer_scientist, researcher
