arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2311.00577stat.MLcs.LGecon.EMstat.ME
收藏

通过正则化和聚类联合分配森林实现多治疗臂的个性化分配

Personalized Assignment to One of Many Treatment Arms via Regularized and Clustered Joint Assignment Forests

University of Michigan(密歇根大学);Stanford University(斯坦福大学);University of Pennsylvania(宾夕法尼亚大学);New York University(纽约大学)

Rahul Ladhania, Jann Spiess, Lyle Ungar, Wenbo Wu

AI总结:

本文针对多治疗臂个性化分配问题,提出正则化森林分配算法并结合聚类方案以跨臂汇集信息,从而在大量治疗臂场景下提升分配效用。

摘要:

我们考虑从随机对照试验中学习针对多个治疗臂之一的个性化分配。由于方差过大,分别对每个治疗臂估计异质性处理效应的标准方法在这种情况下可能表现不佳。我们转而提出跨治疗臂汇集信息的方法:首先,我们考虑一种基于贪婪递归划分的正则化森林分配算法,该算法对各臂之间的效应估计进行收缩。其次,我们通过一种聚类方案增强我们的算法,将结果持续相似的治疗臂合并。在一项模拟研究中,我们将这些方法与分别预测各臂结果的表现进行比较,并记录了通过正则化和聚类直接优化治疗分配所带来的收益。在一个理论模型中,我们说明了大量治疗臂如何使寻找最佳臂变得困难,同时我们能够通过正则化优化从个性化中获得可观的效用收益。

英文摘要:

We consider learning personalized assignments to one of many treatment arms from a randomized controlled trial. Standard methods that estimate heterogeneous treatment effects separately for each arm may perform poorly in this case due to excess variance. We instead propose methods that pool information across treatment arms: First, we consider a regularized forest-based assignment algorithm based on greedy recursive partitioning that shrinks effect estimates across arms. Second, we augment our algorithm by a clustering scheme that combines treatment arms with consistently similar outcomes. In a simulation study, we compare the performance of these approaches to predicting arm-wise outcomes separately, and document gains of directly optimizing the treatment assignment with regularization and clustering. In a theoretical model, we illustrate how a high number of treatment arms makes finding the best arm hard, while we can achieve sizable utility gains from personalization by regularized optimization.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑