It isn’t very useful when using stable or high bias models as base learners.
It’s important to highlight, that we don’t prune decision trees to ensure that low bias and high variance in our base learner.
