Predicting Quantum Monte Carlo Charge Densities using Graph Neural Networks

ORAL

Abstract

The electronic charge density is a fundamental quantity in accurately predicting the properties of

quantum materials. Although density functional approximations have been improving in energetics,

the progress in charge densities has been stagnant. We use quantum Monte Carlo (QMC) to calculate

charge densities as well as total and kinetic energies of over 1700 QM9 dataset molecules with

closed-shell singlet and open-shell triplet configurations. A message-passing graph neural network

designed for grid-based data was trained to predict the charge densities. We explore the viability

of reaching accurate predictions using a small QMC dataset, which we make publicly available for

future machine-learning models.

* This work has been supported by the U.S. Department of Energy, Office of Science, Basic Energy Sciences, Materials Sciences, and Engineering Division, as part of the Computational Materials Sciences Program and Center for Predictive Simulation of Functional Materials. Some part of the research conducted as part of a user project at the Center for Nanophase Materials Sciences (CNMS), which is a US Department of Energy, Office of Science User Facility at Oak Ridge National Laboratory.

–

Presenters

  • Ganesh Panchapakesan

    Oak Ridge National Lab, Oak Ridge National Laboratory

Authors

  • Ganesh Panchapakesan

    Oak Ridge National Lab, Oak Ridge National Laboratory

  • Abdulgani Annaberdiyev

    Oak Ridge National Lab

  • Fan Shu

    Georgia Institute of Technology

  • Victor Fung

    Georgia Institute of Technology