From the course: Applied Machine Learning: Ensemble Learning (2022)

Unlock the full course today

Join today to access over 24,700 courses taught by industry experts.

What is boosting?

What is boosting? - Python Tutorial

From the course: Applied Machine Learning: Ensemble Learning (2022)

Start my 1-month free trial Buy for my team

What is boosting?

“

- [Narrator] Let's dig into our first of three ensemble techniques that we'll be covering in this course. Boosting is an ensemble method that sequentially trains a number of weak models, often trees, to create one strong model. This sounds a lot like our general definition for ensemble learning, but there're two things that I want to emphasize here. The first is that boosting often leverages decision trees as their base models. And the second is that boosting really enforces that these base models are weak. Again, by weak model, I mean one that is only a little better than guessing. And boosting forces these base models to be weak by capping the maximum depth of the trees that make up those weak models. This is an effective way of capping the power of those base models. There's one critically important factor that makes boosting so powerful, and it separates it from most other ensemble learning techniques. That is…

Contents