Dreaming up scale invariance via inverse renormalization group

Abstract

We explore how minimal neural networks can invert the renormalization group (RG) coarse-graining procedure in the two-dimensional Ising model, effectively ``dreaming up'' microscopic configurations from coarse-grained states. This task - formally impossible at the level of configurations - can be approached probabilistically, allowing machine learning models to reconstruct scale-invariant distributions without relying on microscopic input. We demonstrate that even neural networks with as few as three trainable parameters can learn to generate critical configurations, reproducing the scaling behavior of observables such as magnetic susceptibility, heat capacity, and Binder ratios. A real-space renormalization group analysis of the generated configurations confirms that the models capture not only scale invariance but also reproduce nontrivial eigenvalues of the RG transformation. While the inversion is necessarily imperfect, these minimal models robustly reproduce the RG-relevant structure of the critical distribution. Surprisingly, we find that increasing network complexity by introducing multiple layers offers no significant benefit. These findings suggest that simple local rules, akin to those generating fractal structures, are sufficient to encode the universality of critical phenomena, creating an opportunity for efficient generative models of statistical ensembles in physics.

v1: 12 pages, 11 figures, 55 references v2: 13 pages, 11 figures, 61 references

References in corpus (24)

• Machine learning and the physical sciences
• Deep Learning for Single Image Super-Resolution: A Brief Review
• Discovering Phase Transitions with Unsupervised Learning
• Unsupervised learning of phase transitions: from principal component analysis to variational autoencoders
• Machine learning based spatio-temporal super resolution reconstruction of turbulent flows
• Mutual Information, Neural Networks and the Renormalization Group
• Probability distribution of the order parameter for the 3D Ising model universality class: a high precision Monte Carlo study
• AI-assisted super-resolution cosmological simulations
• Super-resolving the Ising model with convolutional neural networks
• Dynamical Regimes of Diffusion Models
• Interpretable machine learning for inferring the phase boundaries in a nonequilibrium system
• Is Deep Learning a Renormalization Group Flow?
• Optimal Renormalization Group Transformation from Information Theory
• Relevance in the Renormalization Group and in Information Theory
• The surprising convergence of the Monte Carlo renormalization group for the d=3 Ising Model
• Statistical physics through the lens of real-space mutual information
• Inverse Renormalization Group based on Image Super-Resolution using Deep Convolutional Networks
• Neural Monte Carlo Renormalization Group
• A Variational Approach to Monte Carlo Renormalization Group
• High-resolution Monte Carlo study of the order-parameter distribution of the three-dimensional Ising model
• Wavelet Conditional Renormalization Group
• Message Passing Variational Autoregressive Network for Solving Intractable Ising Models
• Inverse renormalization group of spin glasses
• Reducing finite-size effects with reweighted renormalization group transformations