Introducing Machine Learning

Chapter 13. How to Make Better Decisions: Ensemble Methods

We need to solve the unsupervised learning problem before we can even think of getting to true AI.

—Yann LeCun(VP and Chief AI Scientist at Facebook)

At the end of the day, the problem of machine learning is always the same: building more and more accurate predictive models. The basic tools available for building solutions are the tools of mathematics we touched on in previous chapters, and the tools to evaluate the quality of the model are bias, variance, and errors. Tree-based algorithms are good for classification problems—which, of a few known labels, is the best to describe the observed data item—but the algorithms presented in the preceding chapter are simple overall and, in some ...