Symbolic Regression with Fast Function Extraction and Nonlinear Least Squares Optimization

Fast Function Extraction (FFX) is a deterministic algorithm for solving symbolic regression problems. We improve the accuracy of FFX by adding parameters to arguments nonlinear functions. Instead only optimizing linear parameters, we optimize these additional with separable least squared optimization using variable projection algorithm. Both and our new applied on PennML benchmark suite. show that proposed extensions leads higher while providing models similar length small increase in runtime given data. Our results are compared large set methods were already published