In this chapter, we introduced artificial neural networks, powerful models for classification and regression that can represent complex functions by composing several artificial neurons. In particular, we discussed directed acyclic graphs of artificial neurons called feedforward neural networks. Multilayer perceptrons are a type of feedforward network in which each layer is fully connected to the subsequent layer. An MLP with one hidden layer and a finite number of hidden units is a universal function approximator. It can represent any continuous function, though it will not necessarily be able to learn appropriate weights automatically. We described how the hidden layers of a network represent latent variables and how their weights can be learned using the backpropagation algorithm. Finally, we used scikit-learn's multilayer perceptron implementation to approximate the function XOR and to classify handwritten digits.

This chapter concludes the book. We discussed a variety of models, learning algorithms, and performance measures, as well as their implementations in scikit-learn. In the first chapter, we described machine learning programs as those that learn from experience to improve their performance at a task. Then, we worked through examples that demonstrated some of the most common experiences, tasks, and performance measures in machine learning. We regressed the prices of pizzas onto their diameters and classified spam and ham text messages. We clustered colors to compress images and clustered the SURF descriptors to recognize photographs of cats and dogs. We used principal component analysis for facial recognition, built a random forest to block banner advertisements, and used support vector machines and artificial neural networks for optical character recognition. Thank you for reading; I hope that you will be able to use scikit-learn and this book's examples to apply machine learning to your own experiences.