what does regularization mean in xgboost (tree)

Question

In xgboost (xgbtree), gamma is the tunning parameter to control the regularization. I understand what regularization means in xgblinear and logistic regression, but in the context of tree-based methods, I'm not sure how regularization works.

Can someone explain how regularization works in xgbtree?

Answer 1

In tree-based methods regularization is usually understood as defining a minimum gain so which another split happens:

Minimum loss reduction required to make a further partition on a leaf node of the tree. The larger gamma is, the more conservative the algorithm will be.

Source: https://xgboost.readthedocs.io/en/latest/parameter.html

This minimum gain can usually be set for anything between $(0,\infty)$.

Here's a somewhat good article on how to tune regularization on XGBoost.