# Physically Interpretable Machine Learning for nuclear masses Source: [Physically Interpretable Machine Learning for nuclear masses]() Collection index: [llms.txt]() Data updated: 2025-10-25T03:44:00.906Z Authors: M. R. Mumpower, T. M. Sprouse, A. E. Lovell, A. T. Mohan Published: 2022-03-20T16:33:50.000Z ## Abstract We present a novel approach to modeling the ground state mass of atomic nuclei based directly on a probabilistic neural network constrained by relevant physics. Our Physically Interpretable Machine Learning (PIML) approach incorporates knowledge of physics by using a physically motivated feature space in addition to a soft physics constraint that is implemented as a penalty to the loss function. We train our PIML model on a random set of $\\sim$20\\% of the Atomic Mass Evaluation (AME) and predict the remaining $\\sim$80\\%. The success of our methodology is exhibited by the unprecedented $\\sigma\_\\textrm{RMS}\\sim186$ keV match to data for the training set and $\\sigma\_\\textrm{RMS}\\sim316$ keV for the entire AME with $Z \\geq 20$. We show that our general methodology can be interpreted using feature importance. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Nuclear mass prediction on AME 2016](): Nuclear mass prediction; AME 2016; IntRMSE: 0.316. - [Nuclear mass excess prediction on AME Experiment I 2020 (interpolation)](): Nuclear mass excess prediction; AME Experiment I 2020 (interpolation); IntRMSE (MeV): 0.246. - [Nuclear mass prediction on AME 2020](): Nuclear mass prediction; AME 2020; ExtRMSE: 0.336. ## Original sources and implementation [Original paper]() [PDF]()