Featureless adaptive optimization accelerates functional electronic materials design

Abstract

Electronic materials exhibiting phase transitions between metastable states (e.g., metal-insulator transition materials with abrupt electrical resistivity transformations) are challenging to decode. For these materials, conventional machine learning methods display limited predictive capability due to data scarcity and the absence of features impeding model training. In this article, we demonstrate a discovery strategy based on multi-objective Bayesian optimization to directly circumvent these bottlenecks by utilizing latent variable Gaussian processes combined with high-fidelity electronic structure calculations for validation in the chalcogenide lacunar spinel family. We directly and simultaneously learn phase stability and band gap tunability from chemical composition alone to efficiently discover all superior compositions on the design Pareto front. Previously unidentified electronic transitions also emerge from our featureless adaptive optimization engine. Our methodology readily generalizes to optimization of multiple properties, enabling co-design of complex multifunctional materials, especially where prior data is sparse.

References in corpus (15)

• Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
• BoltzTraP2, a program for interpolating band structures and calculating semi-classical transport coefficients
• Discovery of low thermal conductivity compounds with first-principles anharmonic lattice dynamics calculations and Bayesian optimization
• Photoferroelectric oxides
• Physical descriptor for the Gibbs energy of inorganic crystalline solids and temperature-dependent materials chemistry
• High-Dimensional Materials and Process Optimization using Data-driven Experimental Design with Well-Calibrated Uncertainty Estimates
• Electric-Field-Induced Resistive Switching in a Family of Mott Insulators : towards Non-Volatile Mott-RRAM Memories
• Ultrafast Structure Switching through Nonlinear Phononics
• Machine learning of phase transitions in the percolation and XY models
• Covalent bonds against magnetism in transition metal compounds
• Electric Pulse Induced Resistive Switching, Electronic Phase Separation, and Possible Superconductivity in a Mott insulator
• Spin-orbital Entangled Molecular $j_{\mathrm{eff}}$ States in Lacunar Spinel Compounds
• First-Order Insulator-to-Metal Mott Transition in the Paramagnetic 3D System GaTa4Se8
• Polar Dynamics at the Jahn-Teller Transition in Ferroelectric GaV4S8
• Assessing exchange-correlation functional performance in the chalcogenide lacunar spinels GaM${}_4$Q${}_8$ (M = Mo, V, Nb, Ta; Q = S, Se)

Cited by in corpus (10)

• Multi-Fidelity Cost-Aware Bayesian Optimization
• Database, Features, and Machine Learning Model to Identify Thermally Driven Metal-Insulator Transition Compounds
• Latent Map Gaussian Processes for Mixed Variable Metamodeling
• Rapid Design of Top-Performing Metal-Organic Frameworks with Qualitative Representations of Building Blocks
• ET-AL: Entropy-Targeted Active Learning for Bias Mitigation in Materials Data
• Uncertainty-Aware Mixed-Variable Machine Learning for Materials Design
• Emerging Microelectronic Materials by Design: Navigating Combinatorial Design Space with Scarce and Dispersed Data
• Mixed-Variable Global Sensitivity Analysis For Knowledge Discovery And Efficient Combinatorial Materials Design
• Energy Landscape analysis of metal-insulator transitions: theory and application to Ca${}_2$RuO${}_4$, $R$NiO${}_3$ and their heterostructures
• Scalable Gaussian Processes for Data-Driven Design using Big Data with Categorical Factors