统计与数据科学系系列学术报告之五百零五期

时间：2026年3月25日（星期三）16:00-17:00

主持人：复旦大学管理学院统计与数据科学系朱仲义教授

地点：史带楼302室

报告人：蒋建成教授大湾区大学

题目：Cross-model mutual learning

摘要：Current transfer learning methods for high-dimensional linear regression assume feature alignment across domains, restricting their applicability to semantically matched features. In many real-world scenarios, however, distinct features in the target and source models can play similar predictive roles, creating a form of cross-model similarity. To leverage this broader transferability, we propose the Cross-model Mutual Learning (CML) framework. It captures potential relationships by comparing each target coefficient with all source coefficients through a weighted fusion penalty. The weights are derived from the derivative of the SCAD penalty, effectively approximating an ideal weighting scheme that preserves transferable signals while filtering out source-specific noise. For computational efficiency, we implement CML using the Alternating Direction Method of Multipliers (ADMM). Theoretically, we establish that under mild conditions, CML achieves the oracle estimator with overwhelming probability. Empirical results from simulations and a real-data application confirm that CML outperforms existing methods in both cross-semantic and partial signal similarity settings.

个人简介：Dr. Jiancheng Jiang returned to China and joined the Great Bay University as a Chair Professor in July 2025. Prior to his return, he served as a joint appointment as a tenured full professor in both the Department of Mathematics and Statistics and the School of Data Science at the University of North Carolina at Charlotte. He also held concurrent roles as the Co-PI of the Charlotte Center for Trustworthy AI and the Statistics Program Coordinator within the Department of Mathematics and Statistics at the same university. After receiving his Ph.D. in Science from Nankai University in 1994, he conducted teaching and research at Peking University, the University of North Carolina at Chapel Hill, and Princeton University. He has served as a Guest Editor for Mathematics, and as an Associate Editor for journals including Statistica Sinica and Frontiers in Artificial Intelligence. He has presided over or participated in more than 10 research projects funded by the U.S. NSF/NIH, and various programs of NSFC (including Young Scientists Fund, General Program, and Key Program). From 2016 to 2020, he was appointed as a Chair Professor at Nankai University. His research focuses on core areas of econometrics, statistics, and machine learning, including distributed computing, financial time series, high-dimensional statistical learning, nonparametric smoothing, quantile regression, and artificial intelligence. He has published more than 70 research papers in peer reviewed journals such as Ann. Statist. and JASA.