A local Bayesian optimizer for atomic structures

Abstract

A local optimization method based on Bayesian Gaussian Processes is developed and applied to atomic structures. The method is applied to a variety of systems including molecules, clusters, bulk materials, and molecules at surfaces. The approach is seen to compare favorably to standard optimization algorithms like conjugate gradient or BFGS in all cases. The method relies on prediction of surrogate potential energy surfaces, which are fast to optimize, and which are gradually improved as the calculation proceeds. The method includes a few hyperparameters, the optimization of which may lead to further improvements of the computational speed.

10 pages, 5 figures

References in corpus (5)

• Machine Learning Unifies the Modelling of Materials and Molecules
• Low-Scaling Algorithm for Nudged Elastic Band Calculations Using a Surrogate Machine Learning Model
• Gaussian Process Regression for Geometry Optimization
• Efficient nonparametric $n$-body force fields from machine learning
• Nudged elastic band calculations accelerated with Gaussian process regression

Cited by in corpus (24)

• Combining Machine Learning and Computational Chemistry for Predictive Insights Into Chemical Systems
• Machine learning for electronically excited states of molecules
• GPAW: An open Python package for electronic-structure calculations
• Perspective on integrating machine learning into computational chemistry and materials science
• Efficient global structure optimization with a machine learned surrogate model
• Predicting the Activity and Selectivity of Bimetallic Metal Catalysts for Ethanol Reforming using Machine Learning
• Global optimization of atomistic structure enhanced by machine learning
• Machine-learning accelerated geometry optimization in molecular simulation
• Global optimization of atomic structures with gradient-enhanced Gaussian process regression
• Atomistic Structure Learning Algorithm with surrogate energy model relaxation
• An Atomistic Machine Learning Package for Surface Science and Catalysis
• Detecting stable adsorbates of (1S)-camphor on Cu(111) with Bayesian optimization
• Enabling robust offline active learning for machine learning potentials using simple physics-based priors
• Machine Learning and Data-Driven Methods in Computational Surface and Interface Science
• ForceNet: A Graph Neural Network for Large-Scale Quantum Calculations
• Atomistic structure search using local surrogate mode
• Atomic structure optimization with machine-learning enabled interpolation between chemical elements
• Machine Learning with bond information for local structure optimizations in surface science
• Electrochemical oxidation of CO on Cu single crystals under alkaline conditions
• Cascading symmetry constraint during machine learning-enabled structural search for sulfur induced Cu(111)-$(\sqrt{43}\times \sqrt{43})$ surface reconstruction
• Accelerating the theoretical study of Li-polysulphide adsorption on single-atom catalysts via machine learning approaches
• Ab initio machine learning in chemical compound space
• Symmetry- and Gradient-enhanced Gaussian Process Regression for the Active Learning of Potential Energy Surfaces in Porous Materials
• Atomistic Global Optimization X: A Python package for optimization of atomistic structures