Hands-On Genetic Algorithms with Python

Deep learning and convolutional neural networks

In recent years, backpropagation algorithms have made a leap forward, enabling the use of a large number of hidden layers in a single network. In these deep neural networks, each layer can interpret a combination of several simpler abstract concepts that were learned by the nodes of the previous layer and produce higher-level concepts. For example, when implementing a face recognition task, the first layer will process the pixels of an image and learn to detect edges in different orientations. The next layer may assemble these into lines, corners, and so on, up to a layer that detects facial features such as nose and lips, and finally, one that combines these into the complete concept of a face. ...