First‐Principles Optical Descriptors and Hybrid Classical–Quantum Classification of Er‐Doped CaF 2

Abstract

ABSTRACT We present a first‐principles‐to‐machine‐learning workflow to classify pristine and Er‐doped using optical descriptors. Finite and clusters derived from fluorite were treated with DFT and LR‐TDDFT in GPAW (PBE, DZP, real‐space finite differences). Excitations up to 10 eV were broadened into continuous spectra, from which transition energy , extinction coefficient , and absorption coefficient were used as features. On the full split, a classical RBF‐SVM reaches 0.983 test accuracy and 0.999 ROC‐AUC. QSVM reaches 0.851 (statevector), 0.817 (noisy simulator), and 0.733 on IBM hardware with reduced 30/15 train/test slices. A hybrid 3‐qubit QNN with a depth‐4 ansatz reaches 0.93 test accuracy and 0.96 AUC. These results show that dopant‐induced spectral fingerprints provide physically interpretable descriptors for benchmarking near‐term quantum models against a strong classical baseline.

More by Dennis Wayo

• Physics‐Guided Neural Regression Modeling of Energy‐Balance Proxies for Hydraulic Fracture Propagation in 1D–3DInternational Journal for Numerical and Analytical Methods in Geomechanics · 2026
• Traversing Gaussian and non-Gaussian regimes in photonic quantum simulators: a reviewEPJ Quantum Technology · 2026
• Decoder Comparability Across Quantum Software Stacks: Repeated-Round Surface and Digitized-GKP Syndrome ReplayarXiv · 2026
• Auditing Syndrome-to-Decoder Mappings for Repetition, Surface, CSS, and Digitized-GKP InterfacesarXiv · 2026