Variable Projection Methods for Solving Separable Nonlinear Inverse Problems

Speaker: Malena Español (Arizona State University)

Date: 9/9/25

Abstract: Variable projection is a powerful and efficient framework for solving separable nonlinear least squares problems, where the model depends linearly on some parameters and nonlinearly on others. In this talk, I will provide an overview of the classical variable projection method and demonstrate its application to large-scale inverse problems, focusing on blind deconvolution. I will also present new regularization strategies and recent extensions of the method that incorporate approximate Jacobians computed using Krylov subspace techniques, which improve efficiency and scalability in challenging inverse problem settings.