AI & Machine Learning · Intermediate

Gradient Boosting

An ensemble technique that builds models sequentially, each one correcting the residual errors of the previous ones.

Definition

Gradient boosting adds weak learners, usually shallow decision trees, that fit the gradient of the loss of the current ensemble. It often achieves state-of-the-art accuracy on structured tabular data. Popular implementations include XGBoost, LightGBM and CatBoost.

Related terms