First-Principles Optical Descriptors and Hybrid Classical-Quantum Classification of Er-Doped CaF$_2$

TL;DR

This paper develops a physics-informed classical-quantum machine learning framework using first-principles optical descriptors to distinguish pristine from Er-doped CaF$_2$, demonstrating high accuracy and robustness of quantum models against classical baselines.

Contribution

It introduces a novel hybrid classical-quantum approach leveraging first-principles optical descriptors for material classification, with detailed DFT and TDDFT calculations and quantum machine learning implementations.

Findings

Classical SVM achieves 98.3% accuracy.

Quantum SVMs reach up to 85.1% accuracy on simulators.

Hybrid quantum neural network attains 93% accuracy.

Abstract

We present a physics-informed classical-quantum machine learning framework for discriminating pristine CaF$_2$ from Er-doped CaF$_2$ using first-principles optical descriptors. Finite Ca$_8$F$_{16}$ and Ca$_7$ErF$_{16}$ clusters were constructed from the fluorite structure (a=5.46~$\AA$) and treated using density functional theory (DFT) and linear-response time-dependent DFT (LR-TDDFT) within the GPAW code. Geometry optimization was performed in LCAO mode with a DZP basis and PBE exchange-correlation functional, followed by real-space finite-difference ground-state calculations with grid spacing h=0.30~$\AA$ and N$_{bands}$=N$_{occ}$+20. Optical excitations up to 10~eV were obtained via the Casida formalism and converted into continuous absorption spectra using Gaussian broadening ($\sigma$=0.1-0.2~eV). From 1,589 energy-resolved points per system, physically interpretable descriptors including transition energy $E$, extinction coefficient $\kappa$, and absorption coefficient $\alpha$ were extracted. A classical RBF-kernel support vector machine (SVM) achieves a test accuracy (ACC) of 0.983 and ROC-AUC of 0.999. Quantum support vector machines (QSVMs) evaluated on statevector and noisy simulators reach accuracies of 0.851 and 0.817, respectively, while execution on IBM quantum hardware yields a test-slice accuracy of 0.733 under finite-shot and decoherence constraints. A hybrid quantum neural network (QNN) with a 3-qubit feature map and depth-4 ansatz achieves a test accuracy of 0.93 and AUC of 0.96. Results here demonstrate that dopant-induced optical fingerprints form a robust, physically grounded feature space for benchmarking near-term quantum learning models against strong classical baselines.