Active learning and element-embedding approach in neural networks for infinite-layer versus perovskite oxides

Combining density functional theory simulations and active learning of neural networks, we explore formation energies of oxygen vacancy layers, lattice parameters, and their statistical correlations in infinite-layer versus perovskite oxides across the periodic table, and place the superconducting nickelate and cuprate families in a comprehensive context. We show that neural networks are capable of predicting these observables with high precision, using only 30-50% of the data for training Element embedding autonomously identifies concepts of chemical similarity between the individual elements that are in line with human knowledge. We demonstrate that active learning efficiently composes the training set by an optimal strategy without a priori knowledge, based on the fundamental concepts of entropy and information, and provides systematic control over the prediction accuracy. This offers key ingredients to considerably accelerate scans of large parameter spaces and exemplifies how artificial intelligence may assist on the quantum scale in finding novel materials with optimized properties.

Automated summary

By combining density functional theory simulations and active learning of neural networks, the study explores the formation energies of oxygen vacancy layers, lattice parameters, and their statistical correlations in different oxides. Neural networks are capable of predicting these observables with high precision, while active learning efficiently composes the training set, accelerating scans of large parameter spaces.