主题:Sequential Quantile Estimation for Distributed and Streaming Data
主讲人:林楠 美国圣路易斯华盛顿大学
主持人:王国长 暨南大学
时间:2025年7月1日(周二)下午16:30-17:30
地点:暨南大学石牌校区经济学院(中惠楼)102室
摘要
Quantile-based methods play a crucial role in understanding distributional effects and capturing heterogeneity in complex systems. However, extending these methods to modern data settings—particularly those involving decentralization or continual arrival of data—presents unique computational and statistical challenges. In this talk, I will discuss recent developments aimed at addressing these challenges, focusing on approaches that enable efficient and scalable quantile inference in distributed and streaming environments. The strategies draw inspiration from ideas such as sequential learning, smoothing techniques, and re-formulations of classic optimization problems. Applications to real-world scenarios and large-scale simulations highlight the practical value of these methods in dynamic and resource-constrained setting.
主讲人简介
林楠教授1999年毕业于中国科学技术大学少年班系,2003年在美国伊利诺伊大学获得统计学专业博士学位,2003-2004年在耶鲁大学做博士后,2004年至今在圣路易斯华盛顿大学任教。现为统计与数据科学系教授。主要从事大数据统计计算、分位数回归,生物信息学以及相关应用领域的研究工作。先后在Biometrika,Biometrics,JCGS,TKDE,New England Journal of Medicine,Genome Research等国际期刊发表70余篇高水平学术论文。曾担任《Computational Statistics & Data Analysis》国际期刊副主编,现任《Journal of Computational and Graphical Statistics》国际期刊副主编。
欢迎感兴趣的师生参加!
校对|王国长
责编| 彭毅
初审| 姜云卢
终审发布| 何凌云
(来源:暨南大学经济学院微信公众号)