Daqian Shao, Ashkan Soleymani, Francesco Quinzan, Marta Kwiatkowska
TL;DR利用双 / 去偏机器学习框架设计的 DML-IV 算法,有效减小两阶段 IV 回归中的偏差并学习高性能策略。
Abstract
A common issue in learning decision-making policies in data-rich settings is
spurious correlations in the offline dataset, which can be caused by hidden
confounders. Instrumental variable (IV) regression, which u