Recent progress on embedded atom neural network approaches

In recent years, atomistic neural network methods for learning potential energy and other physical quantities have advanced rapidly and found widespread applications in chemistry, physics, and materials science.Among these, the embedded atom neural network representation, inspired by the well-known empirical embedded atom method, has facilitated the development of a series of efficient machine learning potentials.This review focuses on the recent progress and applications of the embedded atomic neural network methods and concludes with a discussion of the open challenges and opportunities for the future development of machine learning potentials.