2022年3月28日，加州大学旧金山分校江非助理教授通过线上平台与我中心教员交流，并进行线上学术报告，报告的题目是Time-varying Dynamic Network Model For Dynamic Resting State Functional Connectivity in fMRI and MEG imaging。
近日，清华大学2022年第2期教学简报发布《本科生课程课堂教学质量学生问卷调查统计分析报告（2021-2022学年度秋季学期）》。
清华大学统计学研究中心入围全校前5%的课程如下：
邓婉璐 《初等概率论》
2021-2022学年度秋季学期理论课得分
全校排名第一（100人以上课堂），邓婉璐老师连续两年获此佳绩
周在莹 《统计计算与软件》
2021-2022学年度秋季学期理论课得分
全校排名前5%（100人以上课堂）
周在莹 《非参数统计导论》
2021-2022学年度秋季学期理论课得分
全校排名前5%（30-100人课堂）
此前，在2020-2021学年度春季学期教评中，周在莹老师讲授的两门课《线性回归分析》《实验设计和分析》也排名全校前5%
2022年3月21日，加州大学伯克利分校阮丰博士通过线上平台与我中心教员交流，并进行线上学术报告，报告的题目是Designing Better Nonconvex Model for Modern Statistical Applications。
2022年3月21日，上海财经大学冯兴东教授通过线上平台与我中心教员交流，并进行线上学术报告，报告的题目是Quantile Regression for Nonignorable Missing Data with its Application of Analyzing Electronic Medical Records。
2022年3月8日，哈佛大学生物统计系公布本年度“统计科学领域马文泽伦领导力奖（Marvin Zelen Leadership Award in Statistical Science）”获奖人选，清华大学知名校友、哈佛大学生物统计系林希虹教授获此殊荣，祝贺林老师～也感谢林老师长期以来为清华大学统计学科发展做出的卓越贡献！
Dr. Xihong Lin
Professor of Biostatistics and Coordinating Director of the Program in Quantitative Genomics at the Harvard T.H. Chan School of Public Health and Professor of Statistics at the Faculty of Arts and Sciences of Harvard University, will be the recipient of the 2022 Marvin Zelen Leadership Award in Statistical Science and will deliver a virtual lecture in May 2022.
Described by colleagues as “a force of nature”, “a major star”, and “one of the most influential statistical scientists of our time”, Dr. Lin has distinguished herself as a world leader in statistical methods and applications who has dedicated her career to advancing biostatistics, medicine, and public health.
After earning her PhD at the University of Washington and spending almost 10 years at the University of Michigan, Dr. Lin joined the faculty at Harvard in 2005. Her research has evolved over the years, motivated by the pressing analytical needs in health research. Early in her career, she focused on statistical method development and applications for analysis of longitudinal data and complex observational studies. She became a leader in mixed models, nonparametric and semiparametric regression, missing data, and measurement error for analysis of longitudinal data, and causal inference for complex clinical trials. Later in her career her research shifted to statistical genetics and genomics, with her work becoming a hallmark in statistical methods for analysis of rare variants (RVs) in whole genome sequencing studies. She founded and has been the Coordinating Director of Harvard’s Program in Quantitative of Genomics (PQG) since 2008. Most recently, Dr. Lin has been on the forefront of COVID-19 research, becoming part of a team analyzing the first data out of Wuhan, China and the effectiveness of control measures. In addition to several other efforts, she is the PI of the HowWeFeel project that launched an app in spring 2020 to collect COVID-19 health and exposure data in US and other countries.
Dr. Lin has given back to the statistical and biostatistical community in immeasurable ways through her extraordinary service and leadership. She chaired the Harvard Biostatistics department from 2015-2018, helping to launch an MS program in Health Data Science, one of the first in the nation. As coordinating director of the Program in Quantitative Genomics (PQG), she will help organize its 16th conference, “Emerging Challenges and Opportunities in Gene Editing” later this year. She is the former Chair of the COPSS, and a former member of the Committee of Applied and Theoretical Statistics (CATS) of the National Academy of Science. She is the former Coordinating Editor of Biometrics, the founding co-editor of Statistics in Biosciences, and a former Associate Editor of the Journal of the American Statistical Association and American Journal of Human Genetics.
- Mortimer Spiegelman Award, APHA
- Presidents’ Award, COPSS
- Janet L. Norwood Award for Outstanding Woman Statistician, University of Alabama at Birmingham
- FN David Award, COPSS
- Mosteller Statistician of the Year, ASA Boston Chapter
- Adrienne Cupples Award, Boston University
- Greenberg Distinguished Lectureship Award, UNC Chapel Hill
- Distinguished Alumni Award, University of Washington
- Outstanding Service Award, International Chinese Statistical Association
- Myrto Lefkopoulou Award, Harvard University
- Medallion Lecture Award, Institute of Mathematical Statistics
- Coordinating Director, Program in Quantitative Genomics (PQG)
- Elected Fellow of the American Statistical Association
- Elected Fellow of the Institute of Mathematical Statistics
- Elected to the National Academy of Medicine
- Named one of 50 Changemakers in Public Health by Washington University
- Published over 330 papers, with over 43300 citations
- Principal Investigator or Multiple Principal Investigator on six NIH and NSF grants
- Received the NCI MERIT Award (R37) and the NCI Outstanding Investigator Award
- Chair of the COPSS
- Former member of the Committee of Applied and Theoretical Statistics (CATS) of the National Academy of Science
- Founder, North America Department of Biostatistics Chair Group
- Co-founder, ENAR Young Researcher Workshop
- Co-founder, co-founded the ASA Statistical Genetics and Genomics Section
- Served on the State of Massachusetts COVID-19 Task Force
- Editor, Computational Biology Series
- Founding Co-Editor, Statistics in Biosciences
- Coordinating Editor, Biometrics
- Associate Editor: JASA, AJHG, Statistica, Biometrics, Biometrika, Biostatistics
- Books: Past, Present and Future of Statistical Science and New Developments in Biostatistics and Bioinformatics (Frontiers of Statistics)
以上资料参考自哈佛大学官网：https://www.hsph.harvard.edu/biostatistics/zelenaward/
2022年3月7日，角井（北京）生物技术有限公司周一鸣博士访问我中心，并做学术报告，报告的题目是人工智能技术在抗体药开发中的应用。
近日，2022年第60届国际计算语言学协会年会(Annual Meeting of the Association for Computational Linguistics，简称ACL)举行，我中心邓柯课题组18级博士研究生潘长在、俞声课题组17级博士研究生袁正、18级博士研究生罗声旋几位同学的多篇投稿文章被接收。ACL会议始于1962年，由国际计算语言学协会主办，是自然语言处理与计算语言学领域最高级别的学术会议。
潘长在同学的论文入选“主会长文”单元，题为“ TopWORDS-Seg:开放域中文文本领域通过贝叶斯推断同时进行文本切词和词语发现的方法 (TopWORDS-Seg: Simultaneous Text Segmentation and Word Discovery for Open-Domain Chinese Texts via Bayesian Inference)”，文章针对于几十年来计算语言学中的一个关键瓶颈，开放域中文文本处理问题展开论述。称之为瓶颈是因为在开放域这种具有挑战性的场景中，文本分词和词语发现经常相互纠缠，且并无可用的训练数据。尚无现有方法可以在开放域中同时实现有效的文本分词和单词发现。该文章通过提出一种基于贝叶斯推理的名为 TopWORDS-Seg 的新方法来填补这一空白，在没有训练语料库和领域词表的情况下具有很好的表现和解释性。该文章通过维基百科数据用一系列实验研究证明了 TopWORDS-Seg 的优势。潘长在是第一作者，邓柯副教授作为通讯作者与清华大学计算机系科学与技术系的孙茂松教授共同指导了该工作。
袁正同学共有三篇文章入选：
入选“主会短文”单元文章：“基于疾病同义词的匹配网络的自动疾病编码（Code Synonyms Do Matter: Multiple Synonyms Matching Network for Automatic ICD Coding）”通过额外利用疾病编码的同义词信息去匹配电子病历中的不同文本以达到更好的自动疾病编码效果，在MIMIC-3电子病历数据集上得到了超过以往方法的分类效果。
入选“Findings长文”单元文章 ：“使用三仿射融合异质信息的嵌套命名实体识别方法（Fusing Heterogeneous Factors with Triaffine Mechanism for Nested Named Entity Recognition）”通过三仿射变换改进基于片段分类的命名实体识别模型中的片段表示和片段分类方法，在新闻和医疗命名实体识别任务上取得了超过之前的结果。以上两篇文章袁正均为第一作者，与阿里巴巴达摩院团队合作完成。
此外，袁正与浙江大学、鹏程实验室等研究团队合作的论文：“中文医学自然语言处理评测数据集（CBLUE: A Chinese Biomedical Language Understanding Evaluation Benchmark）也入选了“主会长文”单元。
罗声旋同学的论文入选“Findings长文”单元，题为“联合实体对齐和悬空实体识别的高精度无监督方法 (An Accurate Unsupervised Method for Joint Entity Alignment and Dangling Entity Detection) ”，罗声旋为该文的第一作者，其导师俞声副教授为通讯作者。文章针对在对齐两个知识图谱的现实场景中的三个主要问题：（1）不存在等价对应的实体，也即悬空实体，广泛存在于知识图谱中；（2）悬空实体标签和实体对（等价的两个实体）标签难以获得，一个普适的知识图谱对齐方法需要尽可能避免对监督数据的依赖；（3）各对齐之间以及预测对齐与识别悬空实体之间是互相影响的，需要整体地考虑识别悬空实体并对齐等价的实体。该文章首先根据实体的文本语义信息和全局的相似性指导两个知识图谱中的实体嵌入的训练，从而获得实体之间的距离估计。然后，给每个知识图谱添加一个虚拟实体，从而把实体对齐和悬空实体整合为一个统一的最优运输问题，并解这个问题。最终，与虚拟实体对齐的实体为悬空实体，其余对齐为模型预测的等价实体对。一系列实验表明，该文章在不依赖监督数据的情况下，能够达到当前实体对齐任务上的最优表现，并且有高质量的悬空实体识别结果。
2月24日，在日本京都同志社大学(Doshisha University)举办的第11届国际统计计算协会亚洲分会(The Asian Regional Section of the International Association for Statistical Computing，简称ARS-IASC)会议上，我中心邓柯课题组博士后李艺超博士荣获“杰出年轻学者奖”（Outstanding Young Researcher Award）中的“一等奖”。
ARS-IASC成立于1993年，旨在推进亚太地区统计计算与数据科学的发展。本次会议受到新冠肺炎疫情影响，采用线上线下混合的方式举行，共有来自不同国家和地区的数百人参加。
李艺超通过线上演讲的方式在会议上进行了展示，题目为“序贯蒙特卡洛方法中的分层方法和最优重抽样（Stratification and Optimal Resampling in Sequential Monte Carlo）”，主要对序贯蒙特卡洛中不同情形下的最优重抽样方法进行了解释。
相关成果请参考链接https://academic.oup.com/biomet/article-abstract/109/1/181/6132360。
2021年12月20日，加州大学伯克利分校Martin Wainwright教授通过线上平台为中心师生进行线上特邀报告，报告的题目是Beyond Worst-case: Instance-dependent Optimality in Reinforcement Learning。
2021年12月14日，新加坡国立大学黄东明助理教授通过线上平台为中心师生进行线上学术报告，报告的题目是Controlled Variable Selection with More Flexibility。
