Random Forest extends Bagging by adding an additional source of randomization. What is it?

Question

Answer

Bootstrap samples are replaced by stratified samples to maintain class balance.

Answer

Each tree is trained with a different learning rate to emphasize hard examples.

Answer

Trees are grown to depth 1 (stumps) to ensure weak learners are used.

Answer

At each split, only a random subset of m features (typically m = sqrt(p)) is considered, reducing correlation among trees.

Crowdly