EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Magnetic field reconstruction from proton radiography via physics-guided out-of-distribution filtering

Chun-Sung Jao, Chiung-Yin Chang, Kun-Han Lee, Hsiao-Hsuan Lin, Han Hu, Yen-Chen Chen, Kentaro Sakai, Yao-Li Liu, Po-Yu Chang, Hsiang-Yi Karen Yang, Fang-An Kuo, Tsung-Che Tsai und 1 weitere

APL Machine Learning · 01.09.2026 · 4 (3)

DOI: 10.1063/5.0331075 ↗

Kurz gefasst

Fachartikel in der Zeitschrift APL Machine Learning, online erschienen am 20.08.2026. Verfasst von Chun-Sung Jao und 12 weiteren Autor:innen. Die Arbeit nennt 33 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Institute for Fusion Science, Japan Science and Technology Agency.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Reconstructing multidimensional vector fields from path-integrated projection data is a fundamental challenge in high-energy-density physics, particularly when experimental sources exhibit spectral broadening and shot-to-shot jitter. We present a physics-guided deep-learning framework that addresses this ill-posed inverse problem by formulating global reconstruction as an aggregation of local inference tasks. By training a neural network on single-particle trajectories in randomized uniform magnetic fields, we develop a “local solver” that demonstrates strong zero-shot transfer to previously unseen magnetohydrodynamic topologies. Central to addressing non-ideal laser-driven proton sources, we introduce a spectral out-of-distribution filter that rejects inputs outside the training energy envelope. By preventing extrapolation, the filter enables accurate reconstruction within the represented training domain while maintaining stable populated-cell reconstruction metrics across diverse spectral conditions. Furthermore, we introduce an a priori reconstruction-reliability indicator based on the in-distribution fraction of the source spectrum, which provides a practical estimate of reconstruction coverage before inference. This approach can be integrated with energy-resolved detector systems, such as stacked nuclear track detectors, establishing a practical and highly parallelizable framework for quantitative plasma diagnostics.

Förderung

  • National Institute for Fusion Science
  • Japan Science and Technology Agency

Zitieren

APA

Jao, C., Chang, C., Lee, K., Lin, H., Hu, H., Chen, Y., Sakai, K., Liu, Y., Chang, P., Yang, H. K., Kuo, F., Tsai, T., & Kuramitsu, Y. (2026). Magnetic field reconstruction from proton radiography via physics-guided out-of-distribution filtering. APL Machine Learning, 4(3). https://doi.org/10.1063/5.0331075

Vancouver

Jao C, Chang C, Lee K, Lin H, Hu H, Chen Y, et al. Magnetic field reconstruction from proton radiography via physics-guided out-of-distribution filtering. APL Machine Learning. 2026;4(3). doi:10.1063/5.0331075

BibTeX
@article{Jao2026,
  author = {Jao, Chun-Sung and Chang, Chiung-Yin and Lee, Kun-Han and Lin, Hsiao-Hsuan and Hu, Han and Chen, Yen-Chen and Sakai, Kentaro and Liu, Yao-Li and Chang, Po-Yu and Yang, Hsiang-Yi Karen and Kuo, Fang-An and Tsai, Tsung-Che and Kuramitsu, Yasuhiro},
  title = {Magnetic field reconstruction from proton radiography via physics-guided out-of-distribution filtering},
  journal = {APL Machine Learning},
  year = {2026},
  volume = {4},
  number = {3},
  doi = {10.1063/5.0331075}
}
Kostenlos starten →