题目：Boosted Independence Test for Rare Events with Applications in Vertical Federated Learning

汇报人： 黄丹阳

会议时间：2025年6月5日（周四）16:00

地点：综合楼644会议室

报告人简介：

黄丹阳，中国人民大学亚洲博彩平台排名 教授，吴玉章青年学者，中国人民大学国家治理大数据和人工智能创新平台北京市消费大数据监测子实验室主任。主持国家自然科学基金面上项目、北京市社会科学基金重点项目等科研课题，入选北京市科协青年人才托举工程，曾获北京市优秀人才培养资助。从事网络数据模型、大规模数据计算等方向的理论研究，关注统计理论在中小企业数字化发展中的应用。研究成果三十余篇发表于JRSSB、JASA、JOE、JBES等权威期刊。独著专著《大规模网络数据分析与空间自回归模型》入选“京东统计学图书热卖榜”。获北京高校青年教师教学基本功比赛二等奖、最受学生欢迎奖等多项省部级教学奖励。

报告摘要：

It is generally believed that more observations provide more information. However, we observe that in the independence test for rare events, the power of the test is, surprisingly, determined by the number of rare events rather than the total sample size. We demonstrate this phenomenon in both fixed and high-dimensional settings. To address these issues, we first rescale the covariances to account for the presence of rare events. We then propose a boosted procedure that uses only a small subset of non-rare events, yet achieves nearly the same power as using the full set of observations. As a result, computational complexity is significantly reduced. The theoretical properties, including asymptotic distribution and local power analysis, are carefully derived for both the rescaled statistic based on the full sample and the boosted independent test statistic based on subsampling. We further demonstrate the applicability of our method in vertical federated learning, proposing the Vertical Federated Feature Screening (VFS) algorithm, which effectively reduces computational and communication costs, supported by theoretical guarantees and numerical validations.