To Bag is to Prune
Reviewed by Pith T0 review T1 audit T2 compute T3 formal T4 kernel pith:4BNGPOHErecord.jsonopen to challenge →
read the original abstract
It is notoriously difficult to build a bad Random Forest (RF). Concurrently, RF blatantly overfits in-sample without any apparent consequence out-of-sample. Standard arguments, like the classic bias-variance trade-off or double descent, cannot rationalize this paradox. I propose a new explanation: bootstrap aggregation and model perturbation as implemented by RF automatically prune a latent "true" tree. More generally, randomized ensembles of greedily optimized learners implicitly perform optimal early stopping out-of-sample. So there is no need to tune the stopping point. By construction, novel variants of Boosting and MARS are also eligible for automatic tuning. I empirically demonstrate the property, with simulated and real data, by reporting that these new completely overfitting ensembles perform similarly to their tuned counterparts -- or better.
This paper has not been read by Pith yet.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.