С тех пор произошло много чего интересного (http://venturebeat.com/2015/03/18/google-expert-explains-why-deep-learning-neural-nets-are-hot-in-everything-from-games-to-recognizing-cats/): "In the annual ImageNet competition, the best neural net was able to correctly classify images with a 25.7 percent error rate in 2011. That went to 16.4 percent in 2012, 11.7 percent in 2013, and 6.7 percent in 2014. Baidu showed a paper with a 6.0 percent error rate in January, Microsoft published a 4.9 percent error rate in February, and Google itself published a paper with a 4.8 percent error rate on March 2".
Вот ImageNet -- http://www.image-net.org/
Вот работа Гугля: Using an ensemble of batch-normalized networks, we improve upon the best published result on ImageNet classification: reaching 4.9% top-5 validation error (and 4.8% test error), exceeding the accuracy of human raters -- http://arxiv.org/pdf/1502.03167.pdf