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

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.

arXiv comment4 pages, 5 figures

References in corpus (36)

• Efficient Estimation of Word Representations in Vector Space
• Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
• Deep Potential Molecular Dynamics: a scalable model with the accuracy of quantum mechanics
• New Tolerance Factor to Predict the Stability of Perovskite Oxides and Halides
• A General-Purpose Machine Learning Framework for Predicting Properties of Inorganic Materials
• AFLOW: An automatic framework for high-throughput materials discovery
• Big Data of Materials Science - Critical Role of the Descriptor
• Machine Learning Energies of 2 M Elpasolite (ABC${}_2$D${}_6$) Crystals
• Superconducting Dome in Nd${}_{1-x}$Sr${}_x$NiO${}_2$ Infinite Layer Films
• Quantifying octahedral rotations in strained perovskite oxide films
• A superconducting praseodymium nickelate with infinite layer structure
• Deep Neural Networks for Accurate Predictions of Garnet Stability
• Comparative first-principles studies of prototypical ferroelectric materials by LDA, GGA, and SCAN meta-GGA
• Formation of 2D single-component correlated electron system and band engineering in the nickelate superconductor NdNiO2
• Similarities and differences between infinite-layer nickelates and cuprates and implications for superconductivity
• The critical nature of the Ni spin state in doped NdNiO${}_2$
• Robust $d_{x^2-y^2}$-wave superconductivity of infinite-layer nickelates
• Heterointerface engineered electronic and magnetic phases of NdNiO3 thin films
• Model construction and a possibility of cuprate-like pairing in a new d9 nickelate superconductor (Nd,Sr)NiO2
• Late transition-metal oxides with infinite-layer structure: Nickelates versus cuprates
• Atom2Vec: learning atoms for materials discovery
• Nickelate superconductors -- a renaissance of the one-band Hubbard model
• Topotactic hydrogen in nickelate superconductors and akin infinite-layer oxides ABO2
• Role of $4f$ states in infinite-layer NdNiO${}_2$
• Electronic fitness function for screening semiconductors as thermoelectric materials
• Electronic structure of rare-earth infinite-layer ReNiO2 (Re=La, Nd)
• Hund's metal physics: from SrNiO${}_2$ to NdNiO${}_2$
• Materials design of dynamically stable $d^9$ layered nickelates
• Fundamental difference in the electronic reconstruction of infinite-layer vs. perovskite neodymium nickelate films on SrTiO${}_3$(001)
• Design of $n$- and $p$-type oxide thermoelectrics in LaNiO${}_3$/SrTiO${}_3(001)$ superlattices exploiting interface polarity
• Prediction of thickness limits of ideal polar ultrathin films
• A superlattice approach to doping infinite-layer nickelates
• Confinement- and strain-induced enhancement of thermoelectric properties in LaNiO${}_3$/LaAlO${}_3(001)$ superlattices
• Correlated interface electron gas in infinite-layer nickelate versus cuprate films on SrTiO${}_3$(001)
• Inducing $n$- and $p$-type thermoelectricity in oxide superlattices by strain tuning of orbital-selective transport resonances
• Competition of defect ordering and site disproportionation in strained LaCoO${}_3$ on SrTiO${}_3$(001)