Machine learning of phase transitions in the percolation andXYmodels

2018/04/30 by Wanzhou Zhang, Jiayu Liu, Tzu-Chieh Wei · 115 citations
Mathematics · Physics and Astronomy · #Algorithm #Artificial intelligence #Artificial neural network #Classical XY model #Combinatorics #Complex Network Analysis Techniques #Computer science #Condensed matter physics #Critical exponent #Machine learning #Mathematics #Paramagnetism #Percolation (cognitive psychology) #Phase (matter) #Phase diagram #Phase transition #Physics #Quantum many-body systems #Quantum mechanics #Statistical physics #Theoretical and Computational Physics #Topology (electrical circuits) #cond-mat.dis-nn #cond-mat.quant-gas #cond-mat.stat-mech

paper · pdf · doi:10.1103/physreve.99.032142

published in Physical review. E 99(3), 032142 (American Physical Society) · 13 pages, 16 figures

openalex created_date 2018/04/13 · openalex publication_date 2019/03/29 · arxiv created 2019/06/08 · arxiv updated 2019/06/11 · openalex updated_date 2026/08/06

Abstract

In this paper, we apply machine learning methods to study phase transitions in certain statistical mechanical models on the two-dimensional lattices, whose transitions involve nonlocal or topological properties, including site and bond percolations, the XY model, and the generalized XY model. We find that using just one hidden layer in a fully connected neural network, the percolation transition can be learned and the data collapse by using the average output layer gives correct estimate of the critical exponent ν. We also study the Berezinskii-Kosterlitz-Thouless transition, which involves binding and unbinding of topological defects, vortices and antivortices, in the classical XY model. The generalized XY model contains richer phases, such as the nematic phase, the paramagnetic and the quasi-long-range ferromagnetic phases, and we also apply machine learning method to it. We obtain a consistent phase diagram from the network trained with only data along the temperature axis at two particular parameter Δ values, where Δ is the relative weight of pure XY coupling. Aside from using the spin configurations (either angles or spin components) as the input information in a convolutional neural network, we devise a feature engineering approach using the histograms of the spin orientations in order to train the network to learn the three phases in the generalized XY model and demonstrate that it indeed works. The trained network by using system size L×L can be used to the phase diagram for other sizes (L^'×L^', where L^'≠L) without any further training.

Citations

• Machine learning: Trends, perspectives, and prospects
• A fast Monte Carlo algorithm for site or bond percolation
• Equation of State Calculations by Fast Computing Machines
• Efficient Monte Carlo Algorithm and High-Precision Results for Percolation
• Solving the Quantum Many-Body Problem with Artificial Neural Networks
• Machine learning quantum phases of matter beyond the fermion sign problem
• Pure density functional for strong correlation and the thermodynamic limit from machine learning
• Restricted Boltzmann machines for the long range Ising models
• Unsupervised learning of phase transitions: From principal component analysis to variational autoencoders
• Towards meaningful physics from generative models
• Quantum phase recognition via unsupervised machine learning
• Applications of neural networks to the studies of phase transitions of two-dimensional Potts models

Cited by

• A cautionary tale for machine learning generated configurations in presence of a conserved quantity