报告题目：Statistical Properties of Adversarial Training

报 告 人： 谢宜凌 助理教授 香港城市大学

报告时间：2025年12月10日 16:00-16:40

报告地点：#腾讯会议：448-497-390

校内联系人： 张剑桥 [email protected]

报告摘要：

Motivated by data perturbation, adversarial training has recently been proposed as a new way of parameter estimation in supervised learning. This talk will discuss the statistical properties of the adversarial training estimator from both asymptotic and non-asymptotic perspectives. Firstly, the asymptotic distribution of the adversarial training estimator will be introduced, based on which a new technique has been proposed to improve the performance of existing adversarial training. Secondly, the non-asymptotic convergence rate of the adversarial training estimator will be discussed. The results show that the adversarial training estimator is minimax optimal in high dimensional linear regression.

报告人简介：

谢宜凌，现为香港城市大学决策分析和运营系助理教授。2025年在佐治亚理工娱乐城博士毕业。主要研究最优传输和鲁棒优化的统计性质, 相关成果在 JASA, JMLR, NeurIPS等知名会议期刊上发表。