At a high level, boosting is the process of adding many weak learners to form a strong learner. My professor had mentioned that in the case of boosting using regression trees, trees with a depth of less than 4 or 5 should be appropriate - some sources even say that stubs are sufficiently good weak learners.

Why doesn't boosting make use of multiple strong learners instead; for eg. boosted regression trees using full depth trees?

  • $\begingroup$ @usεr11852 You're right! Didn't see that $\endgroup$
    – rahs
    Dec 26, 2018 at 4:36


Browse other questions tagged or ask your own question.