Congratulations to Prof. Michael Shields on receiving a U.S. Department of Energy (DOE) Early Career Award!
Prof. Shields’s project, titled, “Low-dimensional Manifold Learning for Uncertainty Quantification in Complex Multi-scale Stochastic Systems” leverages large-scale so-called dimension hyper-reduction methods to enable uncertainty quantification for complex multi-scale systems. The advanced modeling approach is likely to be more computationally manageable than conventional methods, allowing for potentially significant impacts in far-ranging fields from computer vision, to language processing, data analysis/machine learning and clustering, and complex networks such as infrastructure and/or communication systems because they afford a fundamental ability to learn from the intrinsic structure of high-dimensional data on the Grassmannian, which is widely recognized as important in these fields. The research developments proposed will also lead to advanced software solutions such as the UQpy open-source Python toolbox for large-scale uncertainty quantification in multiscale stochastic systems.
This research was selected for funding by the Office of Advanced Scientific Computing Research.
The DOE Early Career Research Program, now in its tenth year, is designed to bolster the nation’s scientific workforce by providing support to exceptional researchers during the crucial early career years, when many scientists do their most formative work. To learn more about the award and view information from the other 2019 awardees, click here.