arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1906.09231cs.LGmath.STstat.MLstat.TH
收藏

自适应数据分析经验方法的有效性保证

Guaranteed Validity for Empirical Approaches to Adaptive Data Analysis

Toyota Technical Institute of Chicago(芝加哥丰田理工学院);University of Pennsylvania(宾夕法尼亚大学);Boston University(波士顿大学)

Ryan Rogers, Aaron Roth, Adam Smith, Nathan Srebro, Om Thakkar, Blake Woodworth

摘要:

我们设计了一个用于回答自适应统计查询的通用框架,重点是提供显式置信区间和点估计。在这方面的先前工作集中于为特定分析提供严格的置信区间,或为点估计提供一般的最差情况界限。不幸的是,正如我们所观察到的,这些最坏情况下的界限在许多设置中是松散的-通常甚至没有超过像样本分割这样的简单基线。我们的主要贡献是设计一个框架,提供有效的,特定于实例的置信区间的点估计,可以生成的统计。当与良好的统计学配对时,该方法给出的保证比最佳最坏情况界限好几个数量级。我们提供了一个Python库来实现我们的方法。

英文摘要:

We design a general framework for answering adaptive statistical queries that focuses on providing explicit confidence intervals along with point estimates. Prior work in this area has either focused on providing tight confidence intervals for specific analyses, or providing general worst-case bounds for point estimates. Unfortunately, as we observe, these worst-case bounds are loose in many settings --- often not even beating simple baselines like sample splitting. Our main contribution is to design a framework for providing valid, instance-specific confidence intervals for point estimates that can be generated by heuristics. When paired with good heuristics, this method gives guarantees that are orders of magnitude better than the best worst-case bounds. We provide a Python library implementing our method.

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。

↑