In case you have been living under a rock the last few days, you may have heard that the Association for Computational Machine (ACM) has awarded the 2018 A. M. Turing award to 3 deep learning pioneers, G. Hinton, Y. Bengio, and Y. LeCun. That is the award goes to the people who through their own efforts and those of their students and research associates have made machine learning a household name.
Officially, the ACM has award this prestigious award to a group of people “for conceptual and engineering breakthroughs that have made deep neural networks a critical component of computing.” You can read the rest of the justification at the official website here but essentially these three researchers are responsible of a large number of specialized methods that are so commonplace today that even your manager has heard of them. I’m talking about dropout, back propagation, rectified linear unit, convolutional neural network, and generative adversarial network just to name a few.
Also, I honestly think that G. Hinton together with his students Alex Krizhevsky and Ilya Sutskever are responsible for the GPU revolution in machine learning. Their 2012 paper introducing AlexNet and smashing the competition on the ImageNet challenge is probably the single major event that made deep learning a household name. In the paper, Krizhevsky and his colleagues showed that the power of GPUs can be utilized to get around the computational bottleneck that is training deep neural networks. This paper begun a race among researchers as to who can train the biggest network with the most data on the most GPUs and it resulted in some amazing advances in data-driven machine learning on structured domains such as images, video, audio, and text.
Our humble blog can only say congratulations to the winners and a huge thank you for making machine learning cool and useful.