Randomized algorithms for generalized singular value decomposition with application to sensitivity analysis

The generalized singular value decomposition (GSVD) is a valuable tool that has many applications in computational science. However, computing the GSVD for large-scale problems challenging. Motivated by hyper-differential sensitivity analysis (HDSA), we propose new randomized algorithms which use subspace iteration and weighted QR factorization. Detailed error given provides insight into accuracy of choice algorithmic parameters. We demonstrate performance our on test matrices model problem where HDSA used to study subsurface flow.