Emergence of a finite-size-scaling function in the supervised learning of the Ising phase transition

TL;DR

This paper demonstrates how a simple neural network can learn the universal finite-size-scaling function of the Ising phase transition, accurately predicting critical points across different lattice structures.

Contribution

It introduces a minimal one-parameter neural network model that analytically captures the finite-size-scaling behavior in supervised learning of the Ising transition, validating its universality.

Findings

The model reproduces the finite-size-scaling function in neural network outputs.

It accurately predicts critical points for different lattice types within the same universality class.

The approach is demonstrated on real data from scale-free graphs.

Abstract

We investigate the connection between the supervised learning of the binary phase classification in the ferromagnetic Ising model and the standard finite-size-scaling theory of the second-order phase transition. Proposing a minimal one-free-parameter neural network model, we analytically formulate the supervised learning problem for the canonical ensemble being used as a training data set. We show that just one free parameter is capable enough to describe the data-driven emergence of the universal finite-size-scaling function in the network output that is observed in a large neural network, theoretically validating its critical point prediction for unseen test data from different underlying lattices yet in the same universality class of the Ising criticality. We also numerically demonstrate the interpretation with the proposed one-parameter model by providing an example of finding a critical point with the learning of the Landau mean-field free energy being applied to the real data set from the uncorrelated random scale-free graph with a large degree exponent.