Ceriotti’s perspective on integrated ML models for materials published in MRS Bulletin

By Carey Sargent, EPFL, NCCR MARVEL

MRS Bulletin, a widely recognized publication in advanced materials research, aims to give a comprehensive overview of a specific materials theme topic every month. Written by leading experts, the overview articles are useful references for specialists, but presented at a level that can be understood by a broad scientific audience.   

In October, NCCR MARVEL’s Michele Ceriotti provided an overview of the use of integrated machine learning models for materials. Looking first at how interatomic potentials based on machine learning techniques have become indispensable in the atomic-scale modeling of materials, he also considers future possibilities, pointing out how ML models have been closing the gap with first-principles calculations in predicting arbitrarily complicated functional properties including vibrational and optical spectroscopies and electronic excitations.

“The implementation of integrated ML models that combine energetic and functional predictions with statistical and dynamical sampling of atomic-scale properties is bringing the promise of predictive, uncompromising simulations of existing and novel materials closer to its full realization,” he said in the perspective.

References: 

M. Ceriotti, Beyond potentials: Integrated machine learning models for materials. MRS Bulletin (2022).

DOI: https://doi.org/10.1557/s43577-022-00440-0