Assessing Uncertainty in Nonlinear Inverse Problems with the Metric of Ky Fan

Recently, the metric of Ky Fan was successfully applied to assess uncertainty in linear inverse problems. In this work, we present an extension of these results to nonlinear problems. More precisely, we derive results on convergence and convergence rates for Tikhonov regularization applied to nonlinear stochastic ill-posed problems, and discuss strategies to extend these local convergence results to global ones.