TY - JOUR
T1 - Bayesian optimization of atomic structures with prior probabilities from universal interatomic potentials
AU - Lyngby, Peder
AU - Larsen, Casper
AU - Jacobsen, Karsten Wedel
N1 - Publisher Copyright:
© 2024 American Physical Society.
PY - 2024
Y1 - 2024
N2 - The optimization of atomic structures plays a pivotal role in understanding and designing materials with desired properties. However, conventional computational methods often struggle with the formidable task of navigating the vast potential energy surface, especially in high-dimensional spaces with numerous local minima. Recent advancements in machine learning-driven surrogate models offer a promising avenue for alleviating this computational burden. In this study, we propose anapproach that combines the strengths of universal machine learning potentials with a Bayesian approach using Gaussian processes. By using the machine learning potentials as priors for the Gaussian process, the Gaussian process has to learn only the difference between the machine learning potential and the target energy surface calculated for example by density functional theory. This turns out to improve the speed by which the global optimal structure is identified across diverse systems for a well-behaved machine learning potential. The approach is tested on periodic bulk materials, surface structures, and a cluster.
AB - The optimization of atomic structures plays a pivotal role in understanding and designing materials with desired properties. However, conventional computational methods often struggle with the formidable task of navigating the vast potential energy surface, especially in high-dimensional spaces with numerous local minima. Recent advancements in machine learning-driven surrogate models offer a promising avenue for alleviating this computational burden. In this study, we propose anapproach that combines the strengths of universal machine learning potentials with a Bayesian approach using Gaussian processes. By using the machine learning potentials as priors for the Gaussian process, the Gaussian process has to learn only the difference between the machine learning potential and the target energy surface calculated for example by density functional theory. This turns out to improve the speed by which the global optimal structure is identified across diverse systems for a well-behaved machine learning potential. The approach is tested on periodic bulk materials, surface structures, and a cluster.
U2 - 10.1103/PhysRevMaterials.8.123802
DO - 10.1103/PhysRevMaterials.8.123802
M3 - Journal article
AN - SCOPUS:85211040320
SN - 2476-0455
VL - 8
JO - Physical Review Materials
JF - Physical Review Materials
IS - 12
M1 - 123802
ER -