Inclusive electron-nucleus cross section models from domain adaptation
Abstract
We apply transfer learning (TL) to construct data-driven models of inclusive electron-nucleus cross sections. Starting from an ensemble of deep neural networks pretrained on 12C data, we fine-tune the models separately for 3He, 6Li, 16O, 27Al, 40Ca, and 56Fe. The resulting models improve for all targets, marginally so for oxygen, where the carbon baseline is already adequate, although their predictive robustness depends on the amount, coverage, and precision of the available target data. We systematically study how model performance depends on the number of fine-tuned layers, on the fraction and selection of the training data, and on the overlap between the source and target kinematic domains. The layer-wise analysis shows that oxygen requires only shallow adaptation, whereas helium, calcium, and iron require substantially deeper fine-tuning. Lithium represents the least robust case because of its limited dataset, while aluminum demonstrates a strong sensitivity to a small subset of highly constraining measurements. For selected kinematic configurations outside the coverage of the carbon training data, the adapted models remain consistent with the measurements within their estimated uncertainties. Finally, we compare the resulting predictions with those of the phenomenological F1F2 model.
Related
- Electromagnetic responses induced by one- and two-body nuclear currents in inclusive electron-nucleus scattering
- World average charged-current inclusive neutrino cross section from 6 to 200 GeV
- Domain Adaptation against Background Sculpting in Anomaly Detection at the LHC
- Unsupervised domain adaptation for multitask image analysis in realistic context with extreme label shift; Application to the CTAO first large sized telescope
- Bridging steady-state and time-domain descriptions of molecular electron transport
BibTeX
@misc{graczyk2026inclusive,
title = {{Inclusive electron-nucleus cross section models from domain adaptation}},
author = {Krzysztof M. Graczyk and Beata E. Kowal and Rwik Dharmapal Banerjee and Jose Luis Bonilla and Hemant Prasad and Jan T. Sobczyk},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.08463},
eprint = {2609.08463},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.08463},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Inclusive electron-nucleus cross section models from domain adaptation
AU - Krzysztof M. Graczyk
AU - Beata E. Kowal
AU - Rwik Dharmapal Banerjee
AU - Jose Luis Bonilla
AU - Hemant Prasad
AU - Jan T. Sobczyk
PY - 2026
DA - 2026/09/08
PB - arXiv
DO - 10.48550/arxiv.2609.08463
UR - https://arxiv.org/abs/2609.08463
AB - We apply transfer learning (TL) to construct data-driven models of inclusive electron-nucleus cross sections. Starting from an ensemble of deep neural networks pretrained on ^12C data, we fine-tune the models separately for ^3He, ^6Li, ^16O, ^27Al, ^40Ca, and ^56Fe. The resulting models improve for all targets, marginally so for oxygen, where the carbon baseline is already adequate, although their predictive robustness depends on the amount, coverage, and precision of the available target data. We systematically study how model performance depends on the number of fine-tuned layers, on the fraction and selection of the training data, and on the overlap between the source and target kinematic domains. The layer-wise analysis shows that oxygen requires only shallow adaptation, whereas helium, calcium, and iron require substantially deeper fine-tuning. Lithium represents the least robust case because of its limited dataset, while aluminum demonstrates a strong sensitivity to a small subset of highly constraining measurements. For selected kinematic configurations outside the coverage of the carbon training data, the adapted models remain consistent with the measurements within their estimated uncertainties. Finally, we compare the resulting predictions with those of the phenomenological F1F2 model.
ER -