Gradient boosted decision trees

 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

Добавить комментарий 0

Ваш электронный адрес не будет опубликован. Обязательные поля помечены *