Model reduction for large-scale applications in probabilistic
analysis and inverse problems

Karen Willcox
Department of Aeronautics and Astronautics
Massachusetts Institute of Technology

Model reduction entails the systematic generation of cost-efficient
representations of large-scale systems that result, for example, from
discretization of partial differential equations. Considerable
progress in model reduction methodologies for large-scale systems has
seen successful application to many fields, such as computational
fluid dynamics, structural dynamics, atmospheric modeling, and circuit
design. This talk presents recent developments in methodology to
address the challenges of deriving reduced models that span a wide
range of parametric inputs. Our scalable algorithm for sampling
high-dimensional input spaces makes tractable large-scale
probabilistic analysis and inverse problem applications. The
methodology is demonstrated for inverse problems in contaminant
transport, reacting flow, and porous media flow.