Efficient Prediction of Grain Boundary Energies from Atomistic Simulations via Sequential Design

With the goal of improving data based materials design, it is shown that by a sequential design of experiment scheme the process of generating and learning from the data can be combined to discover the relevant sections of the parameter space. The application is the energy of grain boundaries as a function of their geometric degrees of freedom, calculated from a simple model, or via atomistic simulations. The challenge is to predict the deep cusps of the energy, which are located at irregular intervals of the geometric parameters. Existing sampling approaches either use large sets of datapoints or a priori knowledge of the cusps' positions. By contrast, the authors' technique can find unknown cusps automatically with a minimal amount of datapoints. Key point is a Kriging interpolator with Matern kernel to estimate the energy function. Using the jackknife variance, the next point in the sequential design is a compromise between sampling the region of largest fluctuations and avoiding a clustering of datapoints. In this way, the cusps of the energy can be found within only a few iterations, and refined as desired. This approach will be advantageous for any application with strong, localized fluctuations in the values of the unknown function.

Automated summary

By using a sequential design of experiment scheme, this study proposes a method to combine the generation and learning process of data for improving data-based material design. The application focuses on predicting the energy of grain boundaries based on their geometric degrees of freedom. This technique can automatically discover unknown deep cusps with minimal data points, which is advantageous for applications with strong localized fluctuations in unknown functions.