Second-Order Adjoint Sensitivities: An Alternative Formulation

We present in this work an approach for second-order adjoint sensitivity analysis. Using this approach both the first-and second-order sensitivities of the desired objective function or response are estimated using $n$ extra simulation, where $n$ is the number of parameters. Our approach is simpler than previously developed adjoint approaches as it utilizes Maxwell's equations and not the wave equation. Our approach is illustrated through an example.