About

Rebecca Morrison is a postdoc in the AeroAstro Department at the Massachusetts Institute of Technology, where she works in the group for uncertainty quantification. Her research focus is on sparse graph learning in high-dimensional, non-Gaussian settings, and also the problem of model inadequacy of reduced or uncertain models. These topics are motivated by the goal to make reliable predictions of physical systems using computational, physics-based and data-driven models.

Impact of DiaMonD

As a PhD student with Bob Moser at UT, the support of DiaMonD allowed Rebecca to explore undeveloped areas in uncertainty quantification, especially the effect of uncertain and reduced models. Model error, or inadequacy, is an often overlooked but critical source of error in computational modeling of physical systems. Rebecca’s dissertation addressed this source of error in the kinetics of hydrogen combustion by using data-driven models that respect strict physical constraints. Equally valuable, DiaMonD also provided opportunities to meet researchers from other institutions, including Youssef Marzouk and the UQ group at MIT, where she is now a postdoctoral researcher. This position, in turn, has prepared Rebecca for her upcoming faculty job at CU Boulder.