671. Reducing Bias in Gradient Boosting
medium

A gradient boosting model with shallow trees (depth 2) has high training error after 10 trees. What should be done to reduce bias?