Mechanistic Machine Learning and Digital Twins for Computational Science, Engineering and Technology (MMLDT-CSET) Conference 2021; San Diego, California; September 26-29, 2021

这个为期3天的会议分为14个技术专题和两个混合形式的短期课程。本次会议将提供一个论坛，让学生、研究人员、高中教师和实践者在机械机器学习、人工智能和数字双胞胎领域交流新的想法。应用领域包括土木基础设施、自然灾害工程、地质系统和石油工程、基于可靠性的工程和设计、材料系统、制造业、数学和科学计算、自然和生命科学以及医疗保健。这门名为“机械数据科学”（Mechanistic Data Science， MDS）的短期课程面向高中学生、教师和STEM本科生，旨在为参与者提供与机器学习和数字孪生相关的宏观视角，并通过日常生活中的例子展示如何应用MDS将数据科学工具与数学科学原理结合起来，解决棘手的问题。将为研究生和研究人员开设一门名为“机械机器学习的物理和力学”的短期课程，为具有物理和力学背景的参与者介绍机器学习技术。这些课程还将与“教育、外联和资助机会”技术专题下的指导和交流活动相结合，并在三天的每一天都有一个小组讨论和问答环节。NSF奖学金将用于支持本科生和研究生、博士后、高中教师和学生以及来自贫困学校的学生参加会议和短期课程活动。历史上的黑人学院和大学、少数族裔服务机构以及贫困高中的本科生和研究生将被录取。本次会议介绍了“机械”机器学习和数字孪生（MMLDT）作为将数据与数学和科学原理相结合以解决其他棘手问题的综合方法。本次会议还将“数字双胞胎”确定为通过计算科学、工程和技术（CSET）改进产品设计的重要机器学习应用。MMLDT-CSET的主要目标是通过机械方法和计算科学与工程工具，将这些对学习、开发和应用机器学习和数字孪生感兴趣的不同社区聚集在一起，以解决广泛的工程和科学问题，同时促进工程师、物理和生物科学家、数据和计算机科学家以及来自联邦机构、学术界的数学家之间的跨学科合作。和行业。对未来MMLDT研究和技术发展的讨论将受到社会需求和实践工程师、技术公司和计算机/软件公司提出的重大挑战的推动。该奖项反映了美国国家科学基金会的法定使命，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。

This 3-day conference is organized into fourteen technical tracks and two short courses in a hybrid format. This conference will provide a forum to exchange new ideas among students, researchers, high school teachers, and practitioners in the fields of mechanistic machine learning, artificial intelligence, and digital twins. Application areas include civil infrastructures, natural hazards engineering, geosystems and petroleum engineering, reliability-based engineering and design, material systems, manufacturing, mathematical and scientific computing, natural and life sciences, and healthcare. A short course called “Mechanistic Data Science” (MDS) targeted at high school students and teachers, and STEM undergraduates is designed to provide participants with a big-picture perspective related to machine learning and digital twins, and to demonstrate how to apply MDS to combine data science tools with mathematical scientific principles to solve intractable problems through daily-life examples. A short course called “Mechanistic Machine Learning for Physics and Mechanics” will be offered for graduate students and researchers to introduce machine learning techniques for the participants with a background in physics and mechanics. These courses will also be integrated with the mentoring and networking activities under a Technical Track “Education, Outreach, and Funding Opportunities”, and with a panel and Q&A sessions for each of the three days. NSF Fellowship will be used to support undergraduate and graduate students, postdoctoral fellows, high school teachers and students, as well as students from underprivileged schools to attend the conference and short course activities. Undergraduate and graduate students from historically black colleges and universities and minority-serving institutions as well as underprivileged high schools will be recruited.This conference introduces “Mechanistic” Machine Learning and Digital Twins (MMLDT) as an integrated methodology for coupling data with mathematics and scientific principles to solve otherwise intractable problems. This conference also identifies “Digital Twins” as important machine learning applications for improved product designs via computational science, engineering, and technology (CSET). The main objective of MMLDT-CSET is to bring together these diverse communities that are interested in learning, developing, and applying machine learning and digital twins via mechanistic methods and computational science and engineering tools for a broad range of engineering and scientific problems, while promoting transdisciplinary collaborations among engineers, physical and biological scientists, data and computer scientists, and mathematicians from federal agencies, academia, and industry. The discussion of future MMLDT research and technology developments will be driven by societal needs and grand challenges presented by practicing engineers, technology firms, and computer/software companies.This 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.