计算与应用讨论班

报告题目：Old Is New: Statistical Learning Theory for Contrastive Learning and LLM Alignment

报告人：Yiming Ying（悉尼大学，教授）

时间：2026年7月20日16：00

地点：海纳苑2幢206

摘要：Modern AI changes the learning objects, but many of the underlying statistical questions remain classical. This talk revisits the framework of statistical learning theory—population targets, surrogate calibration, generalization, and approximation—through two examples: contrastive representation learning and large language model alignment. For contrastive learning, we identify the Bayes retrieval score, establish Fisher consistency and calibration of the contrastive logistic loss, and examine how negative sampling and representation approximation affect performance. For Direct Preference Optimization, we introduce contextual preference accuracy as a population alignment target and establish calibration of the DPO logistic objective. We then show that the score-difference parameterization imposes the Bradley–Terry structure: exact representation is equivalent to vanishing triangle curl, while nonzero curl captures an irreducible cyclic component of human preferences. These results illustrate how classical learning theory can clarify both the success and limitations of modern AI objectives.

报告人简介：Dr. Ying is a professor in the School of Mathematics and Statistics at the University of Sydney. Previously, he was a tenured professor in the Department of Mathematics and Statistics at SUNY Albany (USA), where he was also affiliated with Computer Science and founded the UAlbany Machine Learning Group. Dr. Ying received the University at Albany’s Presidential Award for Excellence in Research and Creative Activities (2022) and the SUNY Chancellor’s Award for Excellence in Scholarship and Creative Activities (2023). He regularly serves as a Senior Area Chair for top-tier machine learning conferences including NeurIPS, ICML, AAAI and AISTATS. He currently holds the position of Editor-in-Chief for the Numerical Analysis and Scientific Computation section of Frontiers in Applied Mathematics and Statistics, and works as an associate editor for Analysis and Applications, Machine Learning Journal, and Mathematical Foundation of Computing.

联系人:郭正初（guozc@zju.edu.cn）