Evolutionary identification of nonlinear parametric models with a set-theoretic fitness criterion

System models that are linear in parametric structure, but arbitrarily nonlinear in signal operations, are identified using an approach with two novel components. The fundamental parameter estimation task (the “linear” part) uses a set-theoretic analysis of the data to deduce feasible sets of solutions in light of certain model assumptions. In turn, measurable set solution properties are used to assess the viability of nonlinear regressor functions that compete for “survival” as components of the model best fit to represent the system. The solution is formulated as a somewhat unconventional exercise in evolutionary computation.