Hands-On Gradient Boosting with XGBoost and scikit-learn

Chapter 5: XGBoost Unveiled

In this chapter, you will finally see Extreme Gradient Boosting, or XGBoost, as it is. XGBoost is presented in the context of the machine learning narrative that we have built up, from decision trees to gradient boosting. The first half of the chapter focuses on the theory behind the distinct advancements that XGBoost brings to tree ensemble algorithms. The second half focuses on building XGBoost models within the Higgs Boson Kaggle Competition, which unveiled XGBoost to the world.

Specifically, you will identify speed enhancements that make XGBoost faster, discover how XGBoost handles missing values, and learn the mathematical derivation behind XGBoost's regularized parameter selection. You will establish model templates ...