Exploring machine learning models for predicting meson mass and width

Abstract This paper explores the application of various machine learning models to predict the mass and width of mesons based on their quark content and quantum numbers. Using deep neural networks (DNNs), ensemble methods, and traditional regression models, we demonstrate the effectiveness of these approaches in providing accurate predictions. By utilizing the relationship between isospin, charge conjugation, G-parity, we achieve more precise predictions and reveal the unique differences in meson properties. Additionally, we investigate the effects of incorporating quantum numbers into different ML algorithms, highlighting how these features impact the predictive performance of each model.