Saved in:
Bibliographic Details
Main Authors: Gao, Junjie, Zheng, Xiangyu, Wang, DongDong, Huang, Zhixiang, Zheng, Bangqi, Yang, Kai
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2312.02573
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915141734367232
author Gao, Junjie
Zheng, Xiangyu
Wang, DongDong
Huang, Zhixiang
Zheng, Bangqi
Yang, Kai
author_facet Gao, Junjie
Zheng, Xiangyu
Wang, DongDong
Huang, Zhixiang
Zheng, Bangqi
Yang, Kai
contents Uplift modeling comprises a collection of machine learning techniques designed for managers to predict the incremental impact of specific actions on customer outcomes. However, accurately estimating this incremental impact poses significant challenges due to the necessity of determining the difference between two mutually exclusive outcomes for each individual. In our study, we introduce two novel modifications to the established Gradient Boosting Decision Trees (GBDT) technique. These modifications sequentially learn the causal effect, addressing the counterfactual dilemma. Each modification innovates upon the existing technique in terms of the ensemble learning method and the learning objective, respectively. Experiments with large-scale datasets validate the effectiveness of our methods, consistently achieving substantial improvements over baseline models.
format Preprint
id arxiv_https___arxiv_org_abs_2312_02573
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle UTBoost: Gradient Boosted Decision Trees for Uplift Modeling
Gao, Junjie
Zheng, Xiangyu
Wang, DongDong
Huang, Zhixiang
Zheng, Bangqi
Yang, Kai
Machine Learning
Artificial Intelligence
Uplift modeling comprises a collection of machine learning techniques designed for managers to predict the incremental impact of specific actions on customer outcomes. However, accurately estimating this incremental impact poses significant challenges due to the necessity of determining the difference between two mutually exclusive outcomes for each individual. In our study, we introduce two novel modifications to the established Gradient Boosting Decision Trees (GBDT) technique. These modifications sequentially learn the causal effect, addressing the counterfactual dilemma. Each modification innovates upon the existing technique in terms of the ensemble learning method and the learning objective, respectively. Experiments with large-scale datasets validate the effectiveness of our methods, consistently achieving substantial improvements over baseline models.
title UTBoost: Gradient Boosted Decision Trees for Uplift Modeling
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2312.02573