https://doi.org/10.1140/epja/s10050-025-01710-6
Regular Article - Theoretical Physics
Kolmogorov–Arnold networks for empirical modeling of 
-decay half-lives in superheavy nuclei
Department of Physics, Bharathiar University, 641046, Coimbatore, Tamil Nadu, India
Received:
18
July
2025
Accepted:
29
September
2025
Published online:
1
December
2025
Alpha decay is a dominant decay mode in heavy and superheavy nuclei, offering valuable insight into nuclear structure and stability, particularly in regions with sparse experimental data. In this work, we employ symbolic modeling using Kolmogorov–Arnold networks (KAN) to develop interpretable empirical formulas for predicting -decay half-lives. A dataset of 373 ground-state -emitters was used, with input features derived from both physical properties and prior feature importance analysis via XGBoost. The resulting symbolic expression demonstrates strong agreement with experimental data, achieving a root mean square error (RMSE) of 0.455 and mean absolute error (MAE) of 0.367, while retaining interpretability. Extrapolation tests on superheavy elements with 
–120 further confirm the model’s consistency with existing empirical models, other machine learning models, and experimental values. The findings highlight the effectiveness of KAN in generating compact, accurate, and generalizable models for nuclear decay processes.
Copyright comment Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
© The Author(s), under exclusive licence to Società Italiana di Fisica and Springer-Verlag GmbH Germany, part of Springer Nature 2025
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
