University of Arkansas Spring Lecture Series in the Mathematical Sciences 2019 and 2020

阿肯色大学2019年和2020年春季系列讲座将分别于2019年4月18日至20日和2020年4月16日至18日在AR的费耶特维尔举行，主题为“多元动态系统的贝叶斯分析”和“贝叶斯非参数中的离散随机结构和预测”。这两个系列讲座的共同主题是如何在贝叶斯统计推断的灵活范例中利用复杂模型来利用数据的力量。在每个系列讲座中，该领域的杰出研究人员将针对会议主题进行五次讲座，从介绍性讲座开始，并达到当前研究的边界。此外，会议所涵盖领域的10位领军人物将进行一小时的研究报告，并将有专门的演讲和海报展示环节。研究生、刚获得博士学位的学生和新研究人员都可以获得该奖项。为了方便研究生过渡到将在讲座中讨论的高级研究课题，在会议开始前一天的下午，将有一个专门针对他们的介绍性研讨会。2019年的会议将重点关注贝叶斯方法对高维多元时间序列的建模和分析，广泛涉及统计分析、结构评估、监测和预测等问题。将重点讨论基于解耦/再耦合概念实例化的状态空间模型的最新进展及其隐含策略，这些模型为将连贯的统计分析扩展到日益复杂的动态系统提供了强大的平台。讲座将涵盖基于该概念的最先进的贝叶斯建模方法，以及它们在许多不同领域的广泛应用，包括金融和商业预测，社会经济，工程和自然科学。2020年的会议将概述贝叶斯非参数的最新发展，特别关注离散随机结构，这是许多现代推理目标的关键工具，如主题建模、变化点分析或元分析。将特别关注最有前途的研究方向，如部分可交换性和相关非参数先验，跨越该领域最先进的工具。与会者将有充分的机会建立和促进合作和思想交流，并接触到方法论和跨学科领域的开放研究问题。关于这些会议的更多细节可在https://fulbright.uark.edu/departments/math/research/spring-lecture-series/index.phpThis上获得。奖项反映了NSF的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

The University of Arkansas Spring Lecture Series 2019 and 2020 will be held in Fayetteville, AR, on April 18-20, 2019 and April 16-18, 2020, on the topics of "Bayesian Analysis for Multivariate Dynamic Systems" and "Discrete Random Structures and Prediction in Bayesian Nonparametrics," respectively. The common theme of the two Lecture Series is how to harness the power of data using complex models within the flexible paradigm of Bayesian statistical inference. In each Lecture Series, a prominent researcher in the field will give five lectures on the topic of the conference, starting with an introductory one and reaching the boundaries of current research. Additionally, ten leading figures in the area covered by the conference will give one-hour research presentations, and there will be sessions devoted to contributed talks and poster presentations. Graduate students, recent PhDs, and new researchers are supported by this award. To facilitate the transition of graduate students to the advanced research topics that will be treated in the lectures, there will be an introductory workshop, specifically aimed at them, in the afternoon of the day before the conference begins.The 2019 conference will focus on Bayesian approaches to modeling and analysis of high-dimensional multivariate time series with a broad purview over problems of statistical analysis, structure assessment, monitoring and forecasting. Emphasis will be given on discussion of recent advances in state-space models anchored on instantiations of the decouple/recouple concept and its implied strategies that provide a powerful platform for scaling coherent statistical analysis to increasingly complex dynamic systems. State-of-the-art Bayesian modeling approaches based on this concept, as well as their broad range of applications in many different areas including financial and commercial forecasting, socio-economic, engineering and natural sciences will be covered in the lectures. The 2020 conference will give an overview of recent developments in Bayesian nonparametrics, specifically focusing on discrete random structures that are key tools for many modern inferential goals, such as topic modeling, change-point analyses or meta-analysis. Special attention will be given to the most promising research directions such as partial exchangeability and dependent nonparametric priors, spanning state-of-the-art tools in this area. The attendees will benefit from ample opportunities to forge and foster collaborations and exchange of ideas as well as exposure to open research problems in methodological and cross-disciplinary domains. More details on these conferences are available at https://fulbright.uark.edu/departments/math/research/spring-lecture-series/index.phpThis award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.