Workshop on advancing fluid and soft-matter dynamics with machine learning and data science

This award provides funds to partially support participants in a workshop on advancing fluid and soft-matter dynamics with machine learning and data science. This workshop brings together researchers on these topics to share their research and perspectives on the state of this rapidly evolving area of science and engineering. It also provides a unique opportunity for exchanging ideas between fluid dynamics, soft matter, and machine learning communities, since members of these communities are not necessarily in contact with one another through the normal dissemination venues such as conferences and publications. The participants are from a diverse group of researchers including those from underrepresented groups, early career scientists, and faculty from non-R1 institutions. The structure of the workshop as a small, highly interactive forum with explicit time set aside for discussion will promote cross-fertilization and development of new relationships and collaborations. Recent years have seen an explosion in the use of machine learning and data science tools in Newtonian fluid dynamics, in part due to the availability of software environments for implementing these tools as well as because of improvements in algorithms and computing speed. Relevant applications of machine learning and data science include data-driven closures for RANS models, nonlinear dimension reduction and data-driven time evolution modeling for control applications and combining velocimetry and machine learning to improve velocity field measurements. In soft-matter dynamics, especially non-Newtonian fluid mechanics, machine learning and data science have begun to aid in development of effective constitutive models for very complex soft materials and efficient representations of complex data sets as arise for example in X-ray scattering measurements of complex-fluid microstructure. The workshop will provide a unique forum for researchers across a spectrum of applications, but with common goals and overlapping tools, to learn from one another, and thereby more effectively use machine learning and data science ideas and tools to advance fluid and soft-matter dynamics. 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.