【学术报告】Conformalized Robust Optimization: Model Selection and Robustness Control

报告人：邹长亮（南开大学）

时 间：2025年12月11日16:30

地 点：海韵园行政楼C802

内容摘要:

In decision-making under uncertainty, Contextual Robust Optimization (CRO) provides reliability by minimizing the worst-case decision loss over a prediction set, hedging against label variability. While recent advances use conformal prediction to construct prediction sets for machine learning models, the downstream decisions depend critically on the choice of conformal sets. To address this, we introduce novel model selection framework named Conformalized Robust Optimization with Model Selection (CROMS) that unifies robustness control with decision risk minimization. Furthermore, since traditional coverage is a sufficient but not necessary condition for robustness, enforcing such constraints often leads to overly conservative decisions. To overcome this limitation, we develop a remedy named Conformal Robustness Control (CRC), that directly optimizes the prediction set construction under explicit robustness constraints, thereby enabling more efficient decisions without compromising robustness.

个人简介：

邹长亮，南开大学统计与数据科学学院教授。主要从事统计学及其与数据科学领域的交叉研究和实际应用。研究兴趣包括：预测性推断、高维数据统计学习、变点和异常点检测等。近年来在Ann.Stat.、Biometrika、J.Am.Stat.Asso.、Math. Program等统计学和机器学习领域的权威期刊和会议上发表论文五十余篇，入选爱思唯尔“中国高被引学者”。主持基金委优青、杰青、重点项目、重大项目课题和科技部重点研发计划课题等。任教育部科技委委员、全国应用统计专业硕士教学指导委员会委员、中国现场统计研究会副理事长等。

联系人：周达