Machine learning of phase transitions in the percolation and XY models

TL;DR

This paper demonstrates that machine learning, particularly neural networks, can effectively identify phase transitions and critical points in complex two-dimensional statistical models, including percolation and XY models, using minimal input features.

Contribution

It introduces a neural network approach that accurately detects phase transitions and critical exponents in models with topological phases, using simple input data and feature engineering.

Findings

Neural networks can learn percolation transition with a single hidden layer.

The method accurately estimates the critical exponent $ u$ from data collapse.

The approach successfully maps phase diagrams of complex XY models using minimal data.

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 non-local 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 $\nu$. We also study the Berezinskii-Kosterlitz-Thouless transition, which involves binding and unbinding of topological defects---vortices and anti-vortices, 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 $\Delta$ values, where $\Delta$ is the relative weight of pure XY coupling. Besides 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 $L\times L$ can be used to the phase diagram for other sizes ($L'\times L'$, where $L'\ne L$) without any further training.