Machine-learning force-field models for dynamical simulations of metallic magnets

TL;DR

This paper reviews recent machine learning force-field methods for simulating spin dynamics in metallic magnets, emphasizing scalability, transferability, and the discovery of novel nonequilibrium phenomena.

Contribution

It introduces a symmetry-aware neural network model for efficient, accurate, and transferable spin force predictions in itinerant electron magnets, demonstrated on prototypical models.

Findings

Revealed anomalous coarsening of tetrahedral spin order.

Observed freezing of phase separation in doped systems.

Established ML force-fields as versatile tools for nonequilibrium spin dynamics.

Abstract

We review recent advances in machine learning (ML) force-field methods for Landau-Lifshitz-Gilbert (LLG) simulations of itinerant electron magnets, focusing on scalability and transferability. Built on the principle of locality, a deep neural network model is developed to efficiently and accurately predict the electron-mediated forces governing spin dynamics. Symmetry-aware descriptors constructed through a group-theoretical approach ensure rigorous incorporation of both lattice and spin-rotation symmetries. The framework is demonstrated using the prototypical s-d exchange model widely employed in spintronics. ML-enabled large-scale simulations reveal novel nonequilibrium phenomena, including anomalous coarsening of tetrahedral spin order on the triangular lattice and the freezing of phase separation dynamics in lightly hole-doped, strong-coupling square-lattice systems. These results establish ML force-field frameworks as scalable, accurate, and versatile tools for modeling nonequilibrium spin dynamics in itinerant magnets.