Discovering reduced order model equations of many-body quantum systems using genetic programming: a technical report

In this technical report we present first results of applying Genetic Programming (GP) to obtain equations for constructing reduced order models of quantum systems, with a particular interest in nuclear Density Functional Theory. We employ the reduced basis method to obtain reduced coordinates as the amplitudes of the reduced basis, and use GP to avoid the need of constructing the reduced equations through, for example, a Galerkin projection. The reduced order models constructed through GP show excellent accuracy and speed performance, including extrapolations in the controlling parameters, and show promise as an effective method for emulating computationally demanding calculations in nuclear physics.