Finite-size analysis in neural network classification of critical phenomena

Abstract

We analyze the problem of supervised learning of ferromagnetic phase transitions from the statistical physics perspective. We consider two systems in two universality classes, the two-dimensional Ising model and two-dimensional Baxter-Wu model, and perform careful finite-size analysis of the results of the supervised learning of the phases of each model. We find that the variance of the neural network (NN) output function (VOF) as a function of temperature has a peak in the critical region. Qualitatively, the VOF is related to the classification rate of the NN. We find that the width of the VOF peak displays the finite-size scaling governed by the correlation length exponent, $\nu$, of the universality class of the model. We check this conclusion using several NN architectures -- a fully connected NN, a convolutional NN and several members of the ResNet family -- and discuss the accuracy of the extracted critical exponents $\nu$.

5 pages, 2 figures + Supplementary materials; published version

References in corpus (2)

• Machine-Learning Studies on Spin Models
• Machine-learning detection of the Berezinskii-Kosterlitz-Thouless transitions in the q-state clock models

Cited by in corpus (9)

• Neural density functional theory of liquid-gas phase coexistence
• Why hyperdensity functionals describe any equilibrium observable
• Unsupervised learning of phase transitions via modified anomaly detection with autoencoders
• Learning the bulk and interfacial physics of liquid-liquid phase separation with neural density functionals
• Machine learning of phases and structures for model systems in physics
• Phase determination with and without deep learning
• Influence of anisotropy on the study of critical behavior of spin models by machine learning methods
• Machine Learning Domain Adaptation in Spin Models with Continuous Phase Transitions
• Phase probabilities in first-order transitions using machine learning