Alex Krizhevsky is a machine learning researcher who, as a graduate student at the University of Toronto, built the neural network that triggered the modern deep learning era. Working with his advisor Geoffrey Hinton and fellow student Ilya Sutskever, he designed and trained AlexNet, a deep convolutional neural network described in the 2012 paper “ImageNet Classification with Deep Convolutional Neural Networks.”
AlexNet won the 2012 ImageNet Large Scale Visual Recognition Challenge by a wide margin and demonstrated that deep networks trained on GPUs could dramatically outperform earlier computer vision methods. Krizhevsky’s GPU implementation, released as cuda-convnet, made that training practical, and the result rapidly redirected the entire field toward deep learning. His earlier master’s thesis, “Learning Multiple Layers of Features from Tiny Images,” produced the CIFAR-10 and CIFAR-100 image datasets, which he created with Vinod Nair and Geoffrey Hinton and which remain standard benchmarks for image classification.
After AlexNet, Krizhevsky joined Google. Despite his outsized influence, he keeps a famously low public profile, and his Toronto page still quietly hosts the datasets and code that helped launch the deep learning boom.