株式会社a.s.ist所属の竝河伴裕・林悠偉と、当社技術顧問の岡田真人による論文「GPU-accelerated sequential Monte Carlo for Bayesian spectral analysis」が、米国物理学会（APS）の学術誌「Physical Review E」に2026年8月17日付で採択されました。
論文情報
- 論文タイトル
- GPU-accelerated sequential Monte Carlo for Bayesian spectral analysis
- 著者
- Tomohiro Nabika, Yui Hayashi, Masato Okada
- 掲載誌
- Physical Review E
- 採択日
- 2026年8月17日
Abstract
Bayesian spectral deconvolution provides a data-driven framework for mathematical model selection and parameter estimation from spectral data. Although highly versatile, its computational cost grows with the number of model parameters, data points, and candidate models, often rendering practical applications infeasible.
We propose a GPU-accelerated approach in which a sequential Monte Carlo sampler (SMCS) is executed in parallel on a GPU to perform Bayesian model selection for the number of spectral peaks and Bayesian estimation of peak-function parameters.
Numerical experiments using optimized SIMD CPU baselines demonstrate that the GPU-parallelized SMCS reaches the free-energy accuracy of the established CPU-parallelized replica exchange Monte Carlo (REMC) baseline up to approximately 37× faster—with smaller gains, down to near parity, in the least favorable settings—and is faster than GPU-parallelized REMC in many of the tested cases.
The method is validated on artificial data designed to emulate X-ray photoelectron spectroscopy (XPS) and X-ray diffraction (XRD) measurements, as well as on real experimental spectra.
As measurement techniques such as microscopic spectroscopy and in-situ methods continue to drive rapid growth in the volume of spectral data, the proposed approach offers a practical computational foundation for advanced analysis of each individual dataset.