Gradient boosted decision trees (GBDT) Machine Learning
noun phrase
Definition: An ensemble of decision trees trained sequentially by gradient boosting, where each new tree is fitted to reduce the residual errors (or optimize a differentiable loss) of the existing ensemble [Friedman 2001; scikit-learn/GBDT docs].
Example in context: “Han et al. (2021) used gradient boosted trees and neural network to estimate corporate greenhouse gas emissions for investing decision making.” [Lim 2024]
Synonym: gradient-boosted trees; GBDT; GBDTs; GBT
Related terms: XGBoost, LightGBM