报告题目：A Correlation-Ratio Transfer Learning and Variational Stein's Paradox

报 告 人：林路 教授 山东大学

邀请人：吴婷

报告时间：2022年12月6日(周二) 14:00-15:30

腾讯会议ID：244-943-582

报告人简介：林路，山东大学中泰证券金融研究院教授、博士生导师，第一和第二届教育部应用统计专业硕士教育指导委员会成员，山东省教育厅应用统计专业硕士教育指导委员会成员，山东省政府参事。从事大数据、高维统计、非参数和半参数统计以及金融统计等方面的研究，在国际统计学、机器学习和相关应用学科顶级期刊（包括Ann. Statist., JMLR,《中国科学》）和其它重要期刊发表研究论文120余篇；多个金融资政报告得到省长的正面批示；主持过多项国家自然科学基金课题、全国统计科学研究重大项目、教育部博士点专项基金课题、教育部新文科课题、山东省自然科学基金重点项目等；获得国家统计局颁发的全国统计优秀研究成果一等和二等奖，山东省优秀教学成果一等奖等(均排名第一)。

报告摘要：A basic condition for efficient transfer learning is the similarity between the target model and source models. In practice, however, the similarity condition is difficult to meet or is even violated. Instead of the similarity condition, a brand-new strategy, linear correlation ratio, is introduced in this paper to build an accurate relationship between the models. Such a correlation ratio can be easily estimated by historical data or historical characteristics of permanent variables. Then, an accurate transfer learning likelihood is established based on the correlation ratio combination. On the practical side, the new framework is applied to some application scenarios, specially the area of data streams. Methodologically, some techniques are suggested for transferring the information from simple source models to a relatively complex target model. Theoretically, some favorable properties, including the global convergence rate, are achieved, even for the case where the source models are not similar to the target model. All in all, it can be seen from the theories and experimental results that the inference on the target model is significantly improved by the information from similar or dissimilar source models, and it is somewhat surprising that a related phenomenon of Stein's paradox is illustrated.

主办单位：best365亚洲版登录