Neural-network quantum states for periodic systems in continuous space

We introduce a family of neural quantum states for the simulation of strongly interacting systems in the presence of spatial periodicity. Our variational state is parametrized in terms of a permutationally invariant part described by the Deep Sets neural-network architecture. The input coordinates to the Deep Sets are periodically transformed such that they are suitable to directly describe periodic bosonic systems. We show example applications to both one- and two-dimensional interacting quantum gases with Gaussian interactions, as well as to He-4 confined in a one-dimensional geometry. For the one-dimensional systems we find very precise estimations of the ground-state energies and the radial distribution functions of the particles. In two dimensions we obtain good estimations of the ground-state energies, comparable to results obtained from more conventional methods.

Automated summary

Researchers introduce a family of neural quantum states for simulating strongly interacting systems with spatial periodicity. Their variational state is parametrized by a Deep Sets neural network architecture and includes a permutationally invariant part. By transforming the input coordinates and directly describing periodic bosonic systems, they achieve accurate estimations of ground-state energies and radial distribution functions for one-dimensional systems, as well as comparable results for two-dimensional systems.