Ab initio calculation of real solids via neural network ansatz

TL;DR

This paper introduces a neural network architecture capable of performing ab initio calculations on real solids, demonstrating high accuracy across various systems and paving the way for advanced materials simulations.

Contribution

The authors extend molecular neural networks with periodic boundary conditions, enabling accurate ab initio calculations of real solids using neural network ansatz.

Findings

Outperforms traditional ab initio methods in total energies and cohesive energies

Accurately models electron densities in solids

Applicable to diverse systems like graphene and lithium hydride

Abstract

Neural networks have been applied to tackle many-body electron correlations for small molecules and physical models in recent years. Here we propose a new architecture that extends molecular neural networks with the inclusion of periodic boundary conditions to enable ab initio calculation of real solids. The accuracy of our approach is demonstrated in four different types of systems, namely the one-dimensional periodic hydrogen chain, the two-dimensional graphene, the three-dimensional lithium hydride crystal, and the homogeneous electron gas, where the obtained results, e.g. total energies, dissociation curves, and cohesive energies, outperform many traditional ab initio methods and reach the level of the most accurate approaches. Moreover, electron densities of typical systems are also calculated to provide physical intuition of various solids. Our method of extending a molecular neural network to periodic systems can be easily integrated into other neural network structures, highlighting a promising future of ab initio solution of more complex solid systems using neural network ansatz, and more generally endorsing the application of machine learning in materials simulation and condensed matter physics.