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

TL;DR

This paper combines density functional theory, neural networks, and active learning to efficiently predict properties of oxides, revealing chemical similarities and accelerating materials discovery.

Contribution

It introduces an active learning approach with element embedding for neural networks, enabling accurate predictions with less data and autonomous identification of chemical concepts.

Findings

Neural networks predict formation energies and lattice parameters with high accuracy using only 30-50% of data.

Element embedding captures chemical similarities consistent with human knowledge.

Active learning optimally constructs training sets without prior knowledge, improving prediction control.

Abstract

Combining density functional theory simulations and active learning of neural networks, we explore formation energies of oxygen vacancy layers, lattice parameters, and their correlations in infinite-layer versus perovskite oxides across the periodic table, and place the superconducting nickelate and cuprate families in a comprehensive statistical context. We show that neural networks predict 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 in line with human knowledge. Based on the fundamental concepts of entropy and information, active learning composes the training set by an optimal strategy without a priori knowledge 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.