There is a growing interest in estimating heterogeneous treatment effects across individuals using their high-dimensional feature attributes. Achieving high performance in such high-dimensional heterogeneous treatment effect estimation is challenging because in this setup, it is usual that some features induce sample selection bias while others do not but are predictive of potential outcomes. To avoid losing such predictive feature information, existing methods learn separate feature representations using the inverse of probability weighting (IPW). However, due to the numerically unstable IPW weights, they suffer from estimation bias under a finite sample setup. To develop a numerically robust estimator via weighted representation learning, we propose a differentiable Pareto-smoothed weighting framework that replaces extreme weight values in an end-to-end fashion. Experimental results show that by effectively correcting the weight values, our method outperforms the existing ones, including traditional weighting schemes.

翻译：估计个体间基于高维特征属性的异质性处理效应日益受到关注。在高维异质性处理效应估计场景中，由于部分特征会引发样本选择偏差，而其他特征虽不引发偏差却对潜在结果具有预测能力，因此实现高性能估计颇具挑战性。为避免丢失这类预测性特征信息，现有方法采用逆概率加权（IPW）学习分离的特征表示。然而，受限于IPW权重数值不稳定性，这些方法在有限样本下会产生估计偏差。为通过加权表示学习构建数值鲁棒的估计器，我们提出一种可微帕累托平滑加权框架，该框架能以端到端方式替换极端权重值。实验结果表明，通过有效校正权重值，我们的方法优于包括传统加权方案在内的现有方法。