ADDIS: adaptive algorithms for online FDR control with conservative nulls
ADDIS: adaptive algorithms for online FDR control with conservative nulls
复制标题
ADDIS：具有保守零值的在线 FDR 控制自适应算法
DOI：
--
复制
发表时间：
2019
期刊：
影响因子：
--
通讯作者：
Aaditya Ramdas
中科院分区：
文献类型：
--
作者：
Jinjin Tian;Aaditya Ramdas
大型互联网公司每年例行进行数万次A/B测试。这种大规模的顺序实验导致了最近涌现出的新算法，可以证明可以完全在线方式控制错误发现率(FDR)。然而，如果零点p值是保守的(随机大于均匀分布)，则当前最先进的自适应算法可能遭受显著的功率损失，这种情况在实践中经常发生。在这项工作中，我们引入了一种名为亚的斯亚贝巴的新的自适应丢弃方法，该方法可证明控制FDR并实现了两全其美：如果零值是保守的(实际情况)，它比所有现有方法都有明显的功率增加(实际情况)，而如果零值精确均匀分布(理想情况)，它几乎不会失去功率。我们提供了几个关于调整参数稳健选择的实用见解，并将其扩展到异步和离线设置。
Major internet companies routinely perform tens of thousands of A/B tests each year. Such large-scale sequential experimentation has resulted in a recent spurt of new algorithms that can provably control the false discovery rate (FDR) in a fully online fashion. However, current state-of-the-art adaptive algorithms can suffer from a significant loss in power if null p-values are conservative (stochastically larger than the uniform distribution), a situation that occurs frequently in practice. In this work, we introduce a new adaptive discarding method called ADDIS that provably controls the FDR and achieves the best of both worlds: it enjoys appreciable power increase over all existing methods if nulls are conservative (the practical case), and rarely loses power if nulls are exactly uniformly distributed (the ideal case). We provide several practical insights on robust choices of tuning parameters, and extend the idea to asynchronous and offline settings as well.
影响因子：
6
作者：
Zrnic, Tijana;Ramdas, Aaditya;Jordan, Michael
通讯作者：
Jordan, Michael