International Conference on Mathematics of Data Science

Old Dominion University Research Foundation, Norfolk VA

Investigators

Abstract

Data science is an emerging interdisciplinary field of science and technology. It aims at developing theory, methods and techniques for extraction of useful knowledge or insights from raw data in various structured or unstructured forms such as signal, radar, sounds, images, videos and texts to make smart decisions. Data science employs theories and methods drawn from many fields within the broad areas of mathematics, statistics, information science, and computer science. Mathematics plays an indispensable role in data science. To bring together active researchers in various fields related to data science and practitioners in industry to identify mathematical and statistical challenges in data science, the International Conference on Mathematics of Data Science is being held on the campus of Old Dominion University on November 3-4, 2018. The conference website is http://icmds2018.org. The conference invited speakers are internationally known researchers in the field of data science and will address critical mathematical issues of the field. The conference will promote research collaboration among different areas, cultivate research partnership between academia and industry and, in particular, encourage young talents to work in the field of data science. The funds will solely be used to support junior researchers and graduate students in related fields at US universities and research institutions to attend the conference. This supports the recruitment of young talent to the field of data science and preparation of the next generation researchers to meet the scientific challenges in the big data era. Special efforts are made to recruit graduate students from underrepresented groups such as African American students and female students for the conference participants. The conference will cover mathematical topics crucial to data science. In the field of data science, mathematics has provided functional spaces to represent data sets, approximation approaches to characterize similarity and difference of data sets, optimization methods to extract information from raw data, and analytical, geometrical tools to describe insightful relationships among various concepts in data and their statistical analysis. Further development of data science demands that mathematics play a leading role. For example, it is not yet fully understood that why deep learning is very efficient for certain applications while less efficient in other scenarios. This requires mathematical understanding of the fundamental issues in deep learning. All these issues will be the focus of the conference. Specifically, its scope covers sparse representation of big data sets, functional spaces suitable for big data analysis, mathematical foundation of machine learning, signal image processing, statistical analysis for big data, convex or non-convex sparse optimization for data analysis, scalable algorithms for big data and applications of data science. Scientific and societal broader impacts of this project lie in the aspects that the conference will promote interaction of mathematics, statistics, computer science, engineering and industrial applications, which support the interdisciplinary field of data science, and it will provide a platform for young scholars to learn about and discuss challenging mathematical issues in the field. Website: https://sites.wp.odu.edu/icmds2018/ 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.