论文标题
有界随机变量的接近最佳置信序列
Near-Optimal Confidence Sequences for Bounded Random Variables
论文作者
论文摘要
许多推论问题，例如A/B测试等顺序决策问题，自适应抽样方案（如Bandit Selection）通常在本质上是在线的。在线推论的基本问题是提供一系列置信区间，这些间隔在生长智能样本量的情况下均匀地有效。为了解决这个问题，我们通过利用Bentkus的浓度结果为有限的随机变量提供了近乎最佳的置信序列。我们表明，它改进了使用Cram {é} R-Chernoff技术（例如Hoeffding，Bernstein和Bennett不平等）等现有方法。确认所得的置信序列在合成覆盖问题和适应性停止算法的应用中都很有利。
Many inference problems, such as sequential decision problems like A/B testing, adaptive sampling schemes like bandit selection, are often online in nature. The fundamental problem for online inference is to provide a sequence of confidence intervals that are valid uniformly over the growing-into-infinity sample sizes. To address this question, we provide a near-optimal confidence sequence for bounded random variables by utilizing Bentkus' concentration results. We show that it improves on the existing approaches that use the Cram{é}r-Chernoff technique such as the Hoeffding, Bernstein, and Bennett inequalities. The resulting confidence sequence is confirmed to be favorable in both synthetic coverage problems and an application to adaptive stopping algorithms.
