Machine learning and density functional theory

TL;DR

This paper reviews the progress of machine learning in approximating density functionals within density functional theory, highlighting its potential to transform functional design while questioning if it will replace human expertise.

Contribution

It discusses recent advances in machine learning for functional approximation and explores the implications for future density functional development.

Findings

Machine learning has significantly improved density functional approximations.

The rise of machine learning challenges traditional human-designed functionals.

Future of functional design may increasingly rely on data-driven approaches.

Abstract

Over the past decade machine learning has made significant advances in approximating density functionals, but whether this signals the end of human-designed functionals remains to be seen. Ryan Pederson, Bhupalee Kalita and Kieron Burke discuss the rise of machine learning for functional design.