Which statements about random forests are true?

The choice indicating that each tree is built independently from the others is indeed correct. In a random forest model, multiple decision trees are constructed during the training phase, and each tree is developed without consideration of the other trees that are being created. This independence is critical as it helps in reducing overfitting, which can be a common issue with individual decision trees. Each tree in the random forest learns from a random subset of the data, leading to diverse decision paths and ultimately a more robust model that generalizes better on unseen data.

Moreover, this independence among trees allows the random forest algorithm to aggregate their predictions, typically through majority voting in classification tasks or averaging in regression tasks, which enhances overall predictive performance.