Lesson 6 of 82006–20126 min
ImageNet and AlexNet: data meets GPUs
In 2012 Alex Krizhevsky, a graduate student at the University of Toronto, trained a neural network on a computer with two NVIDIA graphics cards in his bedroom at his parents' house. His co-author was fellow student Ilya Sutskever; their adviser was Geoffrey Hinton.
They entered it in that year's ImageNet Large Scale Visual Recognition Challenge. The network, soon known as AlexNet, got the right label within its top five guesses for all but 15.3 per cent of the test images. The best entry built on other methods had a top-five error of 26.2 per cent. The field noticed.
The rest of this lesson is part of Pro
Create a free account to read the first lesson of every path, and go Pro for the rest.
Sources
- ImageNet Classification with Deep Convolutional Neural Networks (A. Krizhevsky, I. Sutskever, G. Hinton, NIPS 2012) — Advances in Neural Information Processing Systems 25
- ImageNet: A Large-Scale Hierarchical Image Database (J. Deng et al., CVPR 2009) — Princeton University
- CHM Releases AlexNet Source Code — Computer History Museum