NeuralSCF: Neural network self-consistent fields for density functional theory
This work addresses the problem of accelerating electronic structure calculations for computational chemistry and materials science, offering a novel method that enhances accuracy and transferability through mechanics learning.
The authors tackled the computational expense of Kohn-Sham density functional theory (KS-DFT) for large-scale electronic structure calculations by proposing NeuralSCF, a neural network framework that models the Kohn-Sham density map using an SE(3)-equivariant graph transformer, achieving state-of-the-art accuracy in electron density prediction and exceptional zero-shot generalization to out-of-distribution systems.
Kohn-Sham density functional theory (KS-DFT) has found widespread application in accurate electronic structure calculations. However, it can be computationally demanding especially for large-scale simulations, motivating recent efforts toward its machine-learning (ML) acceleration. We propose a neural network self-consistent fields (NeuralSCF) framework that establishes the Kohn-Sham density map as a deep learning objective, which encodes the mechanics of the Kohn-Sham equations. Modeling this map with an SE(3)-equivariant graph transformer, NeuralSCF emulates the Kohn-Sham self-consistent iterations to obtain electron densities, from which other properties can be derived. NeuralSCF achieves state-of-the-art accuracy in electron density prediction and derived properties, featuring exceptional zero-shot generalization to a remarkable range of out-of-distribution systems. NeuralSCF reveals that learning from KS-DFT's intrinsic mechanics significantly enhances the model's accuracy and transferability, offering a promising stepping stone for accelerating electronic structure calculations through mechanics learning.