We highlight that machine-learning interatomic potentials trained over short AIMD trajectories enable first-principles multiscale modeling, bridging DFT level accuracy to the continuum level and empowering the study of complex/novel nanostructures.
Predictive molecular simulations require fast, accurate and reactive interatomic potentials. Machine learning offers a promising approach to construct such potentials by fitting energies and forces to high-level quantum-mechanical data, but...