Abstract
The accuracy of some density functional (DF) models widely used in material science, depends on empirical or free parameters which are commonly tuned using reference physical properties. Grid search methods are the standard numerical approximations used to find the optimal value of the free parameters, making the computational complexity scale with the number of points in the grid. In this report, we illustrate that Bayesian optimization (BO), a sample-efficient machine learning algorithm, can calibrate different density functional models, e.g., hybrid-exchange-correlation and range-separated density functionals. Using the atomization energies and the bond lengths of the Gaussian-1 (G1) and Gaussian-2 (G2) databases, we show that BO optimizes the free parameters of hybrid-exchange-correlation functionals, with approximately 55 evaluations of the error function. We also show that selecting exchange-correlation functionals for different physical systems can be done with BO. We jointly optimize and select the form of the exchange-correlation functional and the free parameters by minimizing the root-mean-square error function for the G1 and G2 datasets atomization energies. The calibrated DF model is on average more accurate than standard DF methods, e.g., PBE0 and B3LYP.
Funding Information
  • Canadian Network for Research and Innovation in Machining Technology, Natural Sciences and Engineering Research Council of Canada

This publication has 55 references indexed in Scilit: