Question

A data scientist is building a tabular churn-prediction model and wants to understand the fundamental difference between two tree-ensemble approaches before choosing one. They plan to run SageMaker training jobs for both a bagging-based method (Random Forest) and a boosting-based method (XGBoost). Which statement correctly describes how these two ensemble methods build their trees?

Accepted Answer

Random Forest builds trees independently in parallel on bootstrapped samples, while XGBoost builds trees sequentially, with each tree correcting the residual errors of the previous ones.

Answer

Both methods build trees sequentially, but XGBoost uses a lower learning rate to prevent overfitting while Random Forest uses a higher one.

Answer

Random Forest builds trees sequentially to reduce bias, while XGBoost builds trees independently to reduce variance.

Answer

Both methods build trees fully independently, and the only difference is that XGBoost prunes leaves while Random Forest does not.