Which ensemble method builds multiple trees by training on bootstrapped samples to ensure diverse decisions?

Which ensemble method builds multiple trees by training on bootstrapped samples to ensure diverse decisions?

• A. Random Forests
• B. Boosting
• C. Gradient Boosting
• D. AdaBoost

Answer: (C)

Random Forests rely on bootstrap sampling to create many diverse decision trees. Each tree is trained on a different bootstrap sample drawn from the training data, so the trees see slightly different data and tend to make different errors. When their predictions are combined—by majority vote for classification or averaging for regression—the ensemble reduces variance and improves generalization compared with a single decision tree. This bagging approach emphasizes diversity among the trees, which is the key to the method's robustness. Boosting methods, by contrast, build trees sequentially to correct previous errors and often use the full dataset for each tree rather than bootstrap samples.

Random Forests rely on bootstrap sampling to create many diverse decision trees. Each tree is trained on a different bootstrap sample drawn from the training data, so the trees see slightly different data and tend to make different errors. When their predictions are combined—by majority vote for classification or averaging for regression—the ensemble reduces variance and improves generalization compared with a single decision tree. This bagging approach emphasizes diversity among the trees, which is the key to the method's robustness. Boosting methods, by contrast, build trees sequentially to correct previous errors and often use the full dataset for each tree rather than bootstrap samples.