Accelerating Finite-temperature Kohn-Sham Density Functional Theory with Deep Neural Networks

TL;DR

This paper introduces a machine learning workflow using deep neural networks to accurately and efficiently reproduce finite-temperature Kohn-Sham DFT total energies, enabling multiscale materials modeling at unprecedented computational scales.

Contribution

The authors develop a neural network-based framework that predicts DFT energies and densities at finite temperature with negligible cost, applicable to both solid and liquid metals.

Findings

Achieves chemical accuracy in total energy predictions

Demonstrates effectiveness for solid and liquid aluminum

Enables multiscale modeling at reduced computational expense

Abstract

We present a numerical modeling workflow based on machine learning (ML) which reproduces the the total energies produced by Kohn-Sham density functional theory (DFT) at finite electronic temperature to within chemical accuracy at negligible computational cost. Based on deep neural networks, our workflow yields the local density of states (LDOS) for a given atomic configuration. From the LDOS, spatially-resolved, energy-resolved, and integrated quantities can be calculated, including the DFT total free energy, which serves as the Born-Oppenheimer potential energy surface for the atoms. We demonstrate the efficacy of this approach for both solid and liquid metals and compare results between independent and unified machine-learning models for solid and liquid aluminum. Our machine-learning density functional theory framework opens up the path towards multiscale materials modeling for matter under ambient and extreme conditions at a computational scale and cost that is unattainable with current algorithms.