时间：7月17日下午：15:00-17:00

地点：主楼418

报告人：张春明

报告人国籍：中国

报告人职称：教授

报告人工作单位：威斯康星大学麦迪逊分校统计系

报告人简介：

Chunming Zhang is a Professor in the Department of Statistics at the University of Wisconsin–Madison. She earned a BS in Mathematical Statistics from Nankai University, an MS in Computational Mathematics from the Chinese Academy of Sciences, and a PhD in Statistics from the University of North Carolina at Chapel Hill. Her research focuses on statistical learning theory and methods applied to computational neuroscience, bioinformatics, and financial econometrics, alongside the analysis of imaging, spatial, and temporal data. Her work also explores dimension reduction and high-dimensional inference, multiple hypothesis testing and large-scale simultaneous inference, nonparametric and semiparametric modeling and inference, functional and longitudinal data analysis, and change-point detection. Dr. Zhang is an elected Fellow of the Institute of Mathematical Statistics (IMS) and the American Statistical Association (ASA), an elected Member of the International Statistical Institute (ISI), and a recipient of the IMS Medallion Award and Lecture (2024).  She has also served on the editorial boards of the Annals of Statistics  and the Journal of the American Statistical Association .  

报告内容简介：

Motivated by inferring causal relationships among neurons using ensemble spike train data, this paper introduces a new technique for learning the structure of a directed acyclic graph (DAG) within a large network of events, applicable to diverse multi-dimensional temporal point process (MuTPP) data. At the core of MuTPP lie the conditional intensity functions, for which we construct a generative model parameterized by the graph parameters of a DAG and develop an equality-constrained estimator, departing from exhaustive search-based methods. We present a novel, flexible augmented Lagrangian (Flex-AL) optimization scheme that ensures provable global convergence and computational efficiency gains over the classical AL algorithm. Additionally, we explore causal structure learning by integrating acyclicity-constraints and sparsity-regularization. We demonstrate: (i) in cases without regularization, the incorporation of the acyclicity constraint is essential for ensuring DAG recovery consistency; (ii) with suitable regularization, the DAG-constrained estimator achieves both parameter estimation and DAG reconstruction consistencies similar to the unconstrained counterpart, but significantly enhances empirical performance. Furthermore, simulation studies indicate that our proposed DAG-constrained estimator, when appropriately penalized, yields more accurate graphs compared to unconstrained or unregularized estimators. Finally, we apply the proposed method to two real MuTPP datasets.

（承办：管理科学与物流系、科研与学术交流中心）