Alex Krizhevsky
Author of AlexNet
The 2012 result that ended the argument
Rating
93Impact score
Domains
Convolutional networksGPU trainingComputer vision
Scouting report
Wrote the CUDA code that trained AlexNet on two consumer GPUs and won ImageNet 2012 by a margin so large it settled the question of whether deep learning worked. Released cuda-convnet so others could reproduce it. Also assembled CIFAR-10 and CIFAR-100, which remain the default first datasets for anyone learning the field.
Curator’s note
Hand-wrote the kernels that started the modern era, then went quiet.
Sources
- 01Alex Krizhevsky en.wikipedia.org
- 02ImageNet Classification with Deep CNNs (2012) proceedings.neurips.cc