Covariate Balancing Methods for Randomized Controlled Trials Are Not Adversarially Robust
Covariate Balancing Methods for Randomized Controlled Trials Are Not Adversarially Robust
复制标题
DOI：
10.1109/tnnls.2023.3266429
复制
发表时间：
2021-10
影响因子：
10.4
通讯作者：
H. Babaei;Sina Alemohammad;Richard Baraniuk
中科院分区：
文献类型：
--
作者：
H. Babaei;Sina Alemohammad;Richard Baraniuk
通过随机试验研究治疗有效性的第一步是将人群分为对照组和治疗组，然后比较接受治疗的治疗组与接受安慰剂的对照组的平均反应。为了确保两组之间的差异仅由治疗引起，控制组和治疗组具有相似的统计数据至关重要。事实上，试验的有效性和可靠性取决于两组统计数据的相似性。协变量平衡方法增加了两组协变量分布之间的相似性。然而，在实践中，往往没有足够的样本来准确估计组的协变量分布。在这篇文章中，我们经验表明，协变量平衡与标准化均值差（SMD）协变量平衡措施，以及Pocock和Simon的序贯治疗分配方法，容易受到最坏情况下的治疗分配。最差情况治疗分配是协变量平衡测量所承认的，但导致最高可能ATE估计误差。我们开发了一种对抗性攻击，以找到任何给定试验的对抗性治疗分配。然后，我们提供了一个指标来衡量给定的试验是如何接近最坏的情况。为此，我们提供了一种基于优化的算法，即治疗效果试验中的对抗性治疗分配（ATASTREET），以找到对抗性治疗分配。
The first step toward investigating the effectiveness of a treatment via a randomized trial is to split the population into control and treatment groups then compare the average response of the treatment group receiving the treatment to the control group receiving the placebo. To ensure that the difference between the two groups is caused only by the treatment, it is crucial that the control and the treatment groups have similar statistics. Indeed, the validity and reliability of a trial are determined by the similarity of two groups’ statistics. Covariate balancing methods increase the similarity between the distributions of the two groups’ covariates. However, often in practice, there are not enough samples to accurately estimate the groups’ covariate distributions. In this article, we empirically show that covariate balancing with the standardized means difference (SMD) covariate balancing measure, as well as Pocock and Simon’s sequential treatment assignment method, are susceptible to worst case treatment assignments. Worst case treatment assignments are those admitted by the covariate balance measure, but result in highest possible ATE estimation errors. We developed an adversarial attack to find adversarial treatment assignment for any given trial. Then, we provide an index to measure how close the given trial is to the worst case. To this end, we provide an optimization-based algorithm, namely adversarial treatment assignment in treatment effect trials (ATASTREET), to find the adversarial treatment assignments.