Estimating nuisance parameters in inverse problems

Many inverse problems include nuisance parameters which, while not of direct interest, are required to recover primary parameters. In this talk, we present the idea of "projecting out" these variables, and how this idea allows us to design methods for solving a broad class of problems with nuisance parameters, such as variance or degrees of freedom. We then discuss several geophysical applications, including including estimation of unknown variance parameters in the Gaussian model for full waveform inversion, degree of freedom (d.o.f.) parameter estimation in the context of robust imaging problems, and robust source estimation.