The Canadian researcher who built AlexNet, the GPU-trained network that won ImageNet in 2012 and set off the modern deep-learning boom.
Alex Krizhevsky is a Canadian computer scientist best known for his work on artificial neural networks and deep learning. In 2012 he, Ilya Sutskever and their PhD advisor Geoffrey Hinton at the University of Toronto built AlexNet, a visual-recognition network trained on just two GeForce GPU cards. It won the ImageNet challenge that year and is widely credited with igniting the deep-learning revolution by showing that deep networks trained on GPUs could far outperform earlier approaches.
Krizhevsky and Sutskever later sold their startup, DNN Research, to Google. He left Google in September 2017 and worked at Dessa on new deep-learning techniques. He is the main author of the CIFAR-10 and CIFAR-100 image datasets, and many of his papers on machine learning and computer vision are highly cited.
With Ilya Sutskever and Geoffrey Hinton, developed a deep convolutional network trained on two GPUs that won the 2012 ImageNet challenge by a wide margin and was presented at NeurIPS.
Was the main author of CIFAR-10 and CIFAR-100, small labelled image datasets that became standard benchmarks in machine learning research.
Together with Ilya Sutskever, sold their startup DNN Research Inc. to Google shortly after AlexNet's debut.