From Complexity to Clarity: Kolmogorov-Arnold Networks in Nuclear Binding Energy Prediction

TL;DR

This paper demonstrates that Kolmogorov-Arnold Networks can effectively predict nuclear binding energies with high accuracy, providing interpretable formulas and surpassing traditional models.

Contribution

It introduces the application of Kolmogorov-Arnold Networks to nuclear physics, achieving improved accuracy and interpretability in binding energy prediction.

Findings

Root mean square error of 0.26 MeV, outperforming traditional models

Derived simplified analytical expressions consistent with classical models

Enhanced interpretability of nuclear binding energy models

Abstract

This study explores the application of Kolmogorov-Arnold Networks (KANs) in predicting nuclear binding energies, leveraging their ability to decompose complex multi-parameter systems into simpler univariate functions. By utilizing data from the Atomic Mass Evaluation (AME2020) and incorporating features such as atomic number, neutron number, and shell effects, KANs achieved a significant lower root mean square error (0.26~MeV), surpassing traditional models. The symbolic regression analysis yielded simplified analytical expressions for binding energies, aligning with classical models like the liquid drop model and the Bethe-Weizs\"acker formula. These results highlight KANs' potential in enhancing the interpretability and understanding of nuclear phenomena, paving the way for future applications in nuclear physics and beyond.