Tenth International Conference on Preconditioning Techniques for Scientific and Industrial Applications

第十届科学和工业应用预处理技术国际会议

• 批准号: 1735572
• 负责人: Yousef Saad
• 金额: $ 1.5万
• 依托单位: University of Minnesota-Twin Cities
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2017
• 资助国家: 美国
• 起止时间: 2017-06-01 至 2018-05-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1735572&HistoricalAwards=false
• 关键词: Tenth; International; Conference; Preconditioning; Techniques

This award supports travel and participation for 15 US-based students and postdoctoral scholars to the 10th International Conference on Preconditioning Techniques for Scientific and Industrial Applications, which will be held July 31st to August 2nd, 2017 at the University of British Columbia in Vancouver, Canada. The conference itself has support from the Pacific Institute for Mathematical Sciences (PIMS); this award is co-funded by NSF CISE CCF and MPS DMS.Scientific and engineering analysis and simulation tasks are often modeled as large systems of equations to be solved; increasingly, these are solved by iterative methods that refine initial guesses. Preconditioning techniques bias the refinements for faster convergence -- they can use quite deep mathematical connections to the sparsity and structure of the matrices that represent the equations. These Preconditioning meetings gather participants from academia, research labs, and industry. (E.g., due to location and topic, a large group are expected from Boeing in Washington.) The resulting mix of participants leads to a healthy exchange of ideas ? with practitioners bringing new, harder, problems to the fore as well as specialized algorithms for handling them, while the algorithms developers and theory researchers emphasize analysis and rigor in what they present. This interaction benefits students who may consider careers in any of these areas.

该奖项支持15名美国学生和博士后学者参加第10届科学和工业应用预处理技术国际会议，该会议将于2017年7月31日至8月2日在加拿大温哥华的不列颠哥伦比亚省大学举行。 会议本身得到了太平洋数学科学研究所（PIMS）的支持;该奖项由NSF CISE CCF和MPS DMS共同资助。科学和工程分析和仿真任务通常被建模为要求解的大型方程组;越来越多地，这些问题通过迭代方法来解决，以改进初始猜测。 预处理技术偏向于更快的收敛--它们可以使用与表示方程的矩阵的稀疏性和结构相当深的数学联系。这些预处理会议聚集了来自学术界、研究实验室和工业界的参与者。(E.g.,由于地点和主题，预计华盛顿波音公司将派出大批人员。）由此产生的参与者的混合导致了健康的思想交流？随着从业者带来新的，更难的问题，以及专门的算法来处理它们，而算法开发人员和理论研究人员强调分析和严谨性，他们提出了什么。这种互动有利于学生谁可能会考虑在任何这些领域的职业生涯。