Login / Signup

Universal scaling in real dimension.

Giacomo BighinTilman EnssNicolò Defenu
Published in: Nature communications (2024)
The concept of universality has shaped our understanding of many-body physics, but is mostly limited to homogenous systems. Here, we present a study of universality on a non-homogeneous graph, the long-range diluted graph (LRDG). Its scaling theory is controlled by a single parameter, the spectral dimension d s , which plays the role of the relevant parameter on complex geometries. The graph under consideration allows us to tune the value of the spectral dimension continuously also to noninteger values and to find the universal exponents as continuous functions of the dimension. By means of extensive numerical simulations, we probe the scaling exponents of a simple instance of O ( N ) symmetric models on the LRDG showing quantitative agreement with the theoretical prediction of universal scaling in real dimensions.
Keyphrases
  • convolutional neural network
  • optical coherence tomography
  • neural network
  • high resolution
  • molecular dynamics
  • machine learning
  • magnetic resonance imaging
  • quantum dots
  • deep learning
  • dual energy
  • living cells