What is the connection between Random Forest and decision tree pruning bias?

Random Forest avoids pruning bias by using fully grown trees, relying instead on averaging for regularization.

Unlike single decision trees that require pruning to avoid overfitting, Random Forest intentionally uses deep, unpruned trees. Overfitting at the tree level is mitigated by averaging across many high-variance models. This shifts regularization from structural simplification (pruning) to ensemble averaging.