Determination of the distribution of strong coupling constant with machine learning

In this work, we use the artificial neural network (ANN) method to study and predict the distribution of strong coupling constants by fitting the existing data. Our approach takes advantage of the ability of ANN to learn complex nonlinear relations and excellent generalization, and allows for a systematic treatment of the uncertainties associated with the data. To ensure the reliability of our results, we apply three evaluation indexes to evaluate the accuracy of model during training. Finally, we obtained the predicted values of the strong coupling constants at different energy scales, and compared and verified them with the existing experimental data. Our approach represents a promising way to improve the determination of the strong coupling constant at low energies, and could have important implications for future experimental and theoretical studies in quantum chromodynamics.