Predicting nucleation using machine learning in the Ising model

We use a Convolutional Neural Network (CNN) and two logistic regression models to predict the probability of nucleation in two-dimensional Ising model. The three successfully for Nearest Neighbor model which classical is observed. CNN outperforms near spinodal Long Range model, but accuracy its predictions decreases as quenches approach spinodal. Occlusion analysis suggests that this decrease due vanishing difference between density nucleating droplet background. Our results are consistent with general conclusion predictability critical point.