Predicting nucleation near the spinodal in the Ising model using machine learning
arXiv:2004.09575 · doi:10.1103/PhysRevE.103.033305
Abstract
We use a Convolutional Neural Network (CNN) and two logistic regression models to predict the probability of nucleation in the two-dimensional Ising model. The three models successfully predict the probability for the Nearest Neighbor Ising model for which classical nucleation is observed. The CNN outperforms the logistic regression models near the spinodal of the Long Range Ising model, but the accuracy of its predictions decreases as the quenches approach the spinodal. Occlusion analysis suggests that this decrease is due to the vanishing difference between the density of the nucleating droplet and the background. Our results are consistent with the general conclusion that predictability decreases near a critical point.
References in corpus (7)
- Adam: A Method for Stochastic Optimization
- Machine learning phases of matter
- Emergent complex neural dynamics
- Discovering Phases, Phase Transitions and Crossovers through Unsupervised Machine Learning: A critical examination
- The structure of fluctuations near mean-field critical points and spinodals and its implication for physical processes
- Zeros of the Partition Function and Pseudospinodals in Long-Range Ising Models
- Prediction in a driven-dissipative system displaying a continuous phase transition