Deep learning of spectra: Predicting the dielectric function of semiconductors

Abstract

Predicting spectra and related properties such as the dielectric function of crystalline materials based on machine learning has a huge, hitherto unexplored, technological potential. For this reason, we create an ab initio database of 9915 dielectric tensors of semiconductors and insulators calculated in the independent-particle approximation (IPA). In addition, we present the OptiMate family of machine learning models, a series of graph attention neural networks (GAT) trained to predict the dielectric function and refractive index. OptiMate yields accurate prediction of spectra of semiconductors using only their crystal structure. Smooth, artifact-free curves are obtained without these properties being enforced by penalties.

References in corpus (20)

• Quantum ESPRESSO: a modular and open-source software project for quantum simulations of materials
• Advanced capabilities for materials modelling with Quantum ESPRESSO
• Optimized norm-conserving Vanderbilt pseudopotentials
• Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
• Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals
• Ab initio Random Structure Searching
• Yambo: an \textit{ab initio} tool for excited state calculations
• A high-bias, low-variance introduction to Machine Learning for physicists
• Atomistic Line Graph Neural Network for Improved Materials Property Predictions
• Recent Progress of the Computational 2D Materials Database (C2DB)
• Many-body perturbation theory calculations using the yambo code
• Machine learning modeling of superconducting critical temperature
• Machine Learning Energies of 2 M Elpasolite (ABC${}_2$D${}_6$) Crystals
• Prediction model of band-gap for AX binary compounds by combination of density functional theory calculations and machine learning techniques
• A critical examination of compound stability predictions from machine-learned formation energies
• Learning the electronic density of states in condensed matter
• Machine-Learning X-ray Absorption Spectra to Quantitative Accuracy
• Prediction of Ambient Pressure Conventional Superconductivity above 80K in Thermodynamically Stable Hydride Compounds
• Modelling the dielectric constants of crystals using machine learning
• SIMPLE code: optical properties with optimal basis functions

Cited by in corpus (3)

• Modelling complex proton transport phenomena -- Exploring the limits of fine-tuning and transferability of foundational machine-learned force fields
• Many-body perturbation theory vs. density functional theory: A systematic benchmark for band gaps of solids
• Discovery of sustainable energy materials via the machine-learned material space