September 17-18, 2019 Albuquerque, NM
Workshop on the Future of Uncertainty Quantification and Multiscale Modeling Across the Department of Energy
Answering scientific questions important to mission areas at the Department of Energy (DOE) requires accurate predictive modeling of complex physical and biological systems along with robust understanding of the uncertainties to guide decision-making. Numerical models describing these systems often involve multiple time and spatial scales as well as complicated coupling conditions. Quantifying uncertainty in the generated predictions requires spanning a high-dimension parameter space. Given that algorithmic and mathematical advances can take decades to move from the initial conceptual phase to the point that they impact scientific discovery, it is important to anticipate and predict future needs for mathematical research directions in these areas.
This workshop aims to bring together early to mid-career researchers in computational science, statistics, computer science, and domain science to discuss the mid-term to long-term (10~20 years) needs and requirements for the development of uncertainty quantification (UQ) and multi-scale modeling techniques that will be important for future advances in many research and application areas at DOE labs. Rather than extending or slightly modifying existing methods in computer science, applied mathematics and statistics, this workshop is focused on the path to sophisticated and systematic development of modeling frameworks in the next decade.
The 1st Applied Math Visioning Workshop was held April 2-3, 2019 at Lawrence Berkeley National Lab and focused on the future of machine learning and data analytics at the Deparment of Energy.