Complex Datasets and Inverse Problems: Tomography, Networks, and Beyond; Rutgers University - New Brunswick, NJ; October 21-22, 2005

复杂数据集和反问题：断层扫描、网络等；

• 批准号: 0534181
• 负责人: Cun-Hui Zhang
• 金额: $ 1.6万
• 依托单位: Rutgers University New Brunswick
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2005
• 资助国家: 美国
• 起止时间: 2005-09-01 至 2006-08-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=0534181&HistoricalAwards=false
• 关键词: Complex; Datasets; Inverse; Problems; Tomography

Abstract Prop ID: DMS-0534181 PI: Zhang, Cun-Hui PO: Shulamith T. Gross Institution: Rutgers University New Brunswick Title: Complex Datasets and Inverse Problems: Tomography, Networks, and Beyond The conference ``Complex Datasets and Inverse Problems: Tomography, Networks, and Beyond'' will be held October 21-22, 2005 at Rutgers University. The conference will focus on a number of important and emerging interdisciplinary areas of research, including medical tomography, networks, and biased data. Statistical tomography algorithms have been playing crucial roles in the development of medical imaging systems, from CAT, PET, SPECT to MRI. In fast functional MRI, brain functions are studied from data sets composed of multiple time series of incomplete Fourier transformation of the deoxy spin density of the brain. Networks are abundant around us: social, energy, traffic, communication, and computer are just some of the examples. Enormous amount of networks data have been collected in the information age we live in, but few statistical tools have been developed for analyzing them as they are typically governed by time-varying and mutually dependent communication protocols sitting on complicated graph-structured network topologies. Many prototypical applications in these and other important technologies can be viewed as statistical inverse problems with large, high-dimensional, and probably biased/incomplete data, which serve as the unifying ground for the conference. The conference will advance several important areas in statistics, including models and methodologies for complex datasets, inverse problems, imaging systems, networks, and incomplete and biased data. Cutting-edge developments of statistical models, methods, and algorithms will be discussed. The conference will have direct impact on a broad range of scientific applications outside the immediate realm of statistics. Examples include functional MRI and other medical imaging systems, telecommunication, energy, transportation, and social networks, network security, bioinformatics, epidemiology, and clinical trials. The conference is expected to attract researchers in different areas of applications, in medical imaging, telecommunications, bio-medical engineering, bioinformatics, epidemiology, and more. These will comprise both internationally renowned experts and graduate students or young researchers who wish to embark in these rapidly progressing interdisciplinary areas. Time will be generously allotted for informal discussion and fruitful exchange of ideas. Through these activities, the conference will play an important role in fostering new research partnership between young and senior participants and among researchers in different areas of applications. The conference will promote research activities, education, and participation of new investigators, graduate students, and researchers from under-represented groups. The proceedings of the conference have been arranged to be published as a volume in the Institute of Mathematical Statistics Monograph series. This publication will help disseminate widely the advances covered in the conference, especially among the researchers who are not able to attend the conference.

摘要项目ID：DMS-0534181 PI：Zhang，Cun-Hui PO：Shulamith T.格罗斯机构：罗格斯大学新玩法标题：复杂数据集和逆问题：断层扫描，网络，和超越 会议“复杂数据集和逆问题：层析成像，网络和超越”将于2005年10月21日至22日在罗格斯大学举行。会议将重点关注一些重要和新兴的跨学科研究领域，包括医学断层扫描，网络和有偏见的数据。统计层析成像算法在从CAT、PET、SPECT到MRI的医学成像系统的发展中起着至关重要的作用。在快速功能MRI中，从由大脑脱氧自旋密度的不完全傅立叶变换的多个时间序列组成的数据集研究大脑功能。网络在我们身边无处不在：社会、能源、交通、通信和计算机只是其中的一些例子。在我们所处的信息时代，已经收集了大量的网络数据，但是很少有统计工具被开发用于分析它们，因为它们通常由位于复杂的图结构网络拓扑上的时变和相互依赖的通信协议来管理。这些和其他重要技术中的许多原型应用可以被视为具有大型，高维，可能有偏见/不完整数据的统计逆问题，这是会议的统一基础。会议将推进统计学的几个重要领域，包括复杂数据集的模型和方法，逆问题，成像系统，网络以及不完整和有偏见的数据。将讨论统计模型、方法和算法的前沿发展。这次会议将对直接统计领域以外的广泛科学应用产生直接影响。示例包括功能性MRI和其他医学成像系统、电信、能源、运输和社交网络、网络安全、生物信息学、流行病学和临床试验。会议预计将吸引不同应用领域的研究人员，包括医学成像，电信，生物医学工程，生物信息学，流行病学等。这些将包括国际知名的专家和研究生或年轻的研究人员谁希望在这些迅速发展的跨学科领域开始。将慷慨地拨出时间进行非正式讨论和富有成果的意见交流。通过这些活动，会议将发挥重要作用，促进青年和高级参与者之间以及不同应用领域的研究人员之间建立新的研究伙伴关系。会议将促进研究活动，教育和新的调查人员，研究生和来自代表性不足的群体的研究人员的参与。会议记录已安排出版的一卷研究所的数理统计专着系列。这份出版物将有助于广泛传播会议所涵盖的进展，特别是在无法参加会议的研究人员中。