Fine-tuning ab initio XANES spectra calculations using the Bayesian optimization algorithm

Area of Science:

• Materials Science
• Computational Chemistry
• Spectroscopy

Background:

• X-ray absorption near-edge structure (XANES) spectroscopy is crucial for understanding material properties.
• Theoretical modeling using FEFF and FDMNES aids in interpreting XANES spectra.
• Parameter optimization for simulations is often complex and time-consuming.

Purpose of the Study:

• To apply Bayesian optimization (BO) for efficient parameter tuning in XANES spectral simulations.
• To compare the performance of BO against random search methods.
• To evaluate different spectrum similarity metrics for assessing spectral agreement.

Main Methods:

• Utilized Bayesian optimization (BO) algorithm for parameter space exploration.
• Employed FEFF and FDMNES simulation codes for theoretical XANES spectra generation.
• Compared BO with random search and analyzed various spectrum similarity metrics.

Main Results:

• Bayesian optimization demonstrated a threefold increase in speed compared to random search.
• Correlation-based similarity metrics yielded superior shape-level agreement over normalized distance metrics.
• Optimized parameters enabled accurate determination of material characteristics.

Conclusions:

• Bayesian optimization offers an efficient and effective approach for tuning XANES simulation parameters.
• The choice of similarity metric significantly impacts the accuracy of spectral comparison.
• This work provides a pathway for faster and more reliable theoretical XANES analysis.