可解释性
Boosting(机器学习)
梯度升压
稳健性(进化)
预测能力
计算机科学
决策树
机器学习
树(集合论)
人工智能
数据集
数据挖掘
数学
随机森林
数学分析
生物化学
化学
哲学
认识论
基因
作者
Anders Hjort,Ida Scheel,Dag Einar Sommervoll,Johan Pensar
标识
DOI:10.1016/j.dss.2023.114106
摘要
We introduce Locally Interpretable Tree Boosting (LitBoost), a tree boosting model tailored to applications where the data comes from several heterogeneous yet known groups with a limited number of observations per group. LitBoost constraints the complexity of a Gradient Boosted Trees model in a way that allows us to express the final model as a set of local Generalized Additive Models, yielding significant interpretability benefits while still maintaining some of the predictive power of a Gradient Boosted Trees model. We use house price prediction as a motivating example and demonstrate the performance of LitBoost on a data set of N=14382 observations from 15 different city districts in Oslo (Norway). We also test the robustness of LitBoost in an extensive simulation study on a synthetic data set.
科研通智能强力驱动
Strongly Powered by AbleSci AI