Development of exchange-correlation functionals assisted by machine learning

TL;DR

This paper reviews how machine learning enhances the development of exchange-correlation functionals in density functional theory, improving accuracy and complementing traditional analytical methods.

Contribution

It provides a comprehensive overview of recent ML applications in functional development, highlighting their role in advancing first-principles calculations.

Findings

ML methods improve exchange-correlation functional accuracy

ML complements analytical approaches in DFT

Recent ML techniques achieve promising performance

Abstract

With the recent rapid progress in the machine-learning (ML), there have emerged a new approach using the ML methods to the exchange-correlation functional of density functional theory. In this chapter, we review how the ML tools are used for this and the performances achieved recently. It is revealed that the ML, not being opposed to the analytical methods, complements the human intuition and advance the development toward the first-principles calculation with desired accuracy.