We present a generic framework for creating differentially private versions of any hypothesis test in a black-box way. We analyze the resulting tests analytically and experimentally. Most crucially, we show good practical performance for small data sets, showing that at epsilon = 1 we only need 5-6 times as much data as in the fully public setting. We compare our work to the one existing framework of this type, as well as to several individually-designed private hypothesis tests. Our framework is higher power than other generic solutions and at least competitive with (and often better than) individually-designed tests.

翻译：我们提出一个通用框架，能以黑盒方式创建任意假设检验的差分隐私版本。通过理论分析与实验验证，我们评估了所提检验方法的性能。最关键的是，我们在小数据集上展现了良好的实际效果：当ε=1时，仅需完全公开场景下5-6倍的数据量即可达到同等检验效能。我们将本工作与现有唯一同类型框架及多个独立设计的隐私假设检验方法进行了对比。结果表明，我们的框架相比其他通用方案具有更高统计功效，且至少与独立设计的检验方法具有竞争力（在多数场景中表现更优）。