4 papers
Quantifying Individual Risk for Binary Outcomes
Peng Wu, Peng Ding, Zhi Geng +1
Understanding treatment effect heterogeneity is crucial for reliable decision-making in treatment evaluation and selection. The conditional average treatment effect (CATE) is widel…
MBD: A Model-Based Debiasing Framework Across User, Content, and Model Dimensions
Yuantong Li, Lei Yuan, Zhihao Zheng +27
Modern recommendation systems rank candidates by aggregating multiple behavioral signals through a value model. However, many commonly used signals are inherently affected by heter…
Pseudo-strata learning via maximizing misclassification reward
Shanshan Luo, Peng Wu, Zhi Geng
Online advertising aims to increase user engagement and maximize revenue, but users respond heterogeneously to ad exposure. Some users purchase only when exposed to ads, while othe…
Safe Individualized Treatment Rules with Controllable Harm Rates
Peng Wu, Qing Jiang, Shanshan Luo +1
Estimating individualized treatment rules (ITRs) is crucial for tailoring interventions in precision medicine. Typical ITR estimation methods rely on conditional average treatment…