The concept of deep learning is not new, as it was introduced to the machine learning domain and to Artificial Neural Networks by Rina Dechter in 1986, and by Igor Aizenberg and colleagues in 2000, respectively. However, it was not until 2012, when the deep learning revolution took place, that some outstanding works having a significant impact on the research community appeared. In connection with computer vision, the AlexNet architecture (the name of the convolutional neural network that the authors designed) was the winner of the ImageNet Large Scale Visual Recognition Challenge (ILSVRC) 2012, with a remarkably low error rate, beating all other competitors by an enormous (15.3% versus 26.2% (second place)) error rate. ImageNet (http://image-net.org/) is a large visual database, containing over 14 million labeled high-resolution images. These images were labeled by humans. ImageNet contains more than 20,000 categories. Therefore, the 2012 breakthrough in solving the ILSVRC 2012 by AlexNet is often considered to be the beginning of the deep learning revolution of the 2010s.