Weak and strong learners

A nice segue (back) into the topic of boosting is a statistical algorithm or model's ability to improve its ability to predict overtime, that is, its performance.

If you are reading this book and have reached this section of this chapter, the assumption is that you understand the concept of machine learning, as it is related to statistical prediction-making. Learning is a computer or model's ability to learn (how to make predictions based upon data) without being explicitly programmed.

We use the term explicitly to mean hardcoded selections based upon data values.

If you build upon this concept, you can come to understand that a computer or model whose intention or objective is to make good predictions (to guess an ...

Get Statistics for Data Science now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.