Application of interpretable data-driven methods for the reconstruction of supernova neutrino energy spectra following fast neutrino flavor conversions

The simulation of fast flavor conversions under realistic astrophysi cal conditions requires substantial computational resources and involves significant analytical chal lenges. While machine learning methods like Multilayer Perceptrons have been used to accurately predict the asymptotic outcomes of FFCs, their ‘black-box’ nature limits the extraction of direct physical insight.

To mitigate this limitation, we employ two distinct interpretable machine learning frameworks—Kolmogorov-Arnold Networks (KANs) and Sparse Identification of Nonlinear Dynam ics (SINDy)—to learn interpretable surrogates for the asymptotic input–output mapping from a FFC simulation dataset. Our analysis reveals a fundamental trade-off between predictive accu racy and model simplicity. The KANs demonstrates high fidelity in reconstructing post-conversion neutrino energy spectra, achieving accuracies of up to 90%.