Generative modeling of nucleon-nucleon interactions

Developing high-precision models of the nuclear force and propagating the associated uncertainties in quantum many-body calculations of nuclei and nuclear matter remain primary challenges for ab initio nuclear theory. In the present work we develop a generative machine learning model for constructing nuclear potentials that will allow for a more complete estimation of statistical uncertainties arising from the arbitrary choice of nuclear interaction and resolution scale. We train the generative model on nucleon-nucleon potentials derived from chiral effective field theory at three different choices of the resolution scale. We then show that the model can be used to generate novel samples of the nucleon-nucleon potential drawn from a continuous distribution in the resolution scale parameter space. The generated potentials are shown to produce high-quality nucleon-nucleon scattering phase shifts.