Orbital-free Bond Breaking via Machine Learning

TL;DR

This paper demonstrates that machine learning can effectively approximate the kinetic energy functional for diatomic molecules, enabling accurate bond dissociation and potential applications in molecular dynamics.

Contribution

The study introduces a machine learning-based approach to approximate the kinetic energy functional, improving accuracy and systematic trainability for diatomic molecules.

Findings

Accurate dissociation of diatomic molecules achieved

Self-consistent densities and forces closely match reference data

Potential for ab-initio molecular dynamics simulations

Abstract

Machine learning is used to approximate the kinetic energy of one dimensional diatomics as a functional of the electron density. The functional can accurately dissociate a diatomic, and can be systematically improved with training. Highly accurate self-consistent densities and molecular forces are found, indicating the possibility for ab-initio molecular dynamics simulations.