Skip to content
Open
No due date
Last updated Nov 8, 2021
25% complete

When agtboost handles

  • Training on influence-adjusted derivatives
  • Optimized L2-tuning from adjusted information-criterion
  • Automatic stochastic variations overt trees
  • Smart internal feature-engineering with categorical features

The resulting trained models should be extremely close to optimal gradient-tree-boosting ensembles

List view