Learning Full Configuration Interaction Electron Correlations with Deep Learning

TL;DR

This paper introduces eCPNN, a deep learning framework that learns compact electron correlation potentials to accurately predict atomic energies, offering a new approach to modeling complex electron interactions.

Contribution

The paper presents a novel deep learning method, eCPNN, capable of learning electron correlation potentials from limited data to accurately predict atomic energies.

Findings

eCPNN accurately predicts FCI energies for studied atoms.

eCPNN learns succinct potential functions describing electron correlations.

Unsupervised training with limited data is effective for complex quantum systems.

Abstract

In this report, we present a deep learning framework termed the Electron Correlation Potential Neural Network (eCPNN) that can learn succinct and compact potential functions. These functions can effectively describe the complex instantaneous spatial correlations among electrons in many--electron atoms. The eCPNN was trained in an unsupervised manner with limited information from Full Configuration Interaction (FCI) one--electron density functions within predefined limits of accuracy. Using the effective correlation potential functions generated by eCPNN, we can predict the total energies of each of the studied atomic systems with a remarkable accuracy when compared to FCI energies.