Critical behavior of the Ising model in annealed scale-free networks

Sang Hoon Lee, Meesoon Ha, Hawoong Jeong, Jae Dong Noh, Hyunggyu ParkPhysical 2009 · DOI

TL;DR

This paper investigates the critical behavior of the Ising model on annealed scale-free (SF) networks, revealing a complex, multi-regime finite-size scaling (FSS) behavior that depends on network structure and a forced degree cutoff. The findings suggest that this "anomalous FSS" is a general feature of critical phenomena in annealed complex networks, extending beyond non-equilibrium models.

What Problem Does This Paper Address?

Understanding phase transitions in complex systems often relies on studying simplified models on idealized structures. While the Ising model is a cornerstone for studying equilibrium phase transitions, its behavior on highly heterogeneous networks, like scale-free (SF) networks, is less understood, especially concerning finite-size effects. Standard finite-size scaling (FSS) theories, developed for regular lattices, assume a single characteristic length scale. However, SF networks, with their broad degree distributions, introduce complexities. This paper specifically addresses how the FSS of the Ising model behaves on annealed SF networks, where network structure itself can fluctuate. This is contrasted with quenched networks where the structure is fixed. The authors aim to understand if the "anomalous FSS" observed in non-equilibrium models on annealed SF networks also appears in equilibrium models like the Ising model, and how network properties like degree distribution and a forced upper cutoff influence this behavior.

Methods

The core of the paper involves mapping the Ising model on an annealed SF network to a weighted fully connected Ising model. This transformation allows for analytical derivation of the free energy and subsequent FSS. The authors then develop an FSS theory tailored for SF networks, considering the impact of a forced upper cutoff in the degree distribution. This theory predicts multiple scaling regimes and dependence on network parameters. To validate their analytical results, the authors perform extensive Monte Carlo (MC) simulations of the Ising model on annealed SF networks. They specifically focus on comparing their analytical predictions for the order parameter and susceptibility with simulation data across various network parameters (degree exponent $\lambda$, cutoff exponent $\omega$) and system sizes ($N$).

Datasets

Not discussed in the paper. The study uses simulated data generated through Monte Carlo simulations based on the described model.

Key Insights and Ideas

  • Anomalous FSS in Equilibrium Models: The most significant insight is that the "anomalous FSS" previously observed in non-equilibrium models (like the contact process) on annealed SF networks also manifests in the equilibrium Ising model. This suggests that the underlying mechanism for this complex scaling behavior is a general property of annealed complex networks, not specific to non-equilibrium dynamics.
  • Two-Parameter Scaling: The FSS is not governed by a single scaling variable but by two, leading to multiple distinct scaling regimes. This contrasts with standard FSS theory on regular lattices.
  • Cutoff Dependence: The presence of a forced upper cutoff in the degree distribution ($k_c$) plays a crucial role. The FSS behavior and the number of scaling regimes are directly influenced by this cutoff and its relation to the degree exponent ($\omega$ vs. $\lambda$).
  • Universality vs. Specificity: While bulk critical exponents might be universal, the FSS behavior on annealed SF networks is shown to be highly specific, depending on the network's structural properties (degree distribution, cutoff) and the annealing process.
  • Discrepancy Between Annealed and Quenched: The paper reinforces the observation that annealed and quenched SF networks exhibit different FSS behaviors, highlighting the importance of network dynamics in critical phenomena.

Key Results

  • Analytic Derivation of FSS: The paper provides analytic expressions for the finite-size scaling functions of the order parameter and susceptibility for the Ising model on annealed SF networks.
  • Multiple Scaling Regimes: For SF networks with $3 < \lambda < 5$, the FSS exhibits up to four distinct scaling regimes, characterized by different scaling variables and exponents, depending on the cutoff exponent $\omega$.
  • Parameter Dependence: The critical FSS exponents and scaling functions are found to depend on the degree exponent $\lambda$ and the cutoff exponent $\omega$.
  • Validation by Simulation: Monte Carlo simulations confirm the analytic predictions for the scaling behavior of the order parameter across different parameter regimes, showing good agreement with the derived FSS forms.
  • Non-Self-Averaging Property: For certain parameter ranges ($3 < \lambda < 5$ and $\omega = \omega_{nat}$), the order parameter is shown to be non-self-averaging, meaning its value fluctuates significantly between different network samples, a phenomenon visually confirmed by histograms.

Limitations and Future Work

  • Focus on Specific SF Networks: The analysis primarily focuses on SF networks with a power-law degree distribution and a forced upper cutoff. The behavior of other network topologies or degree distributions is not explored.
  • Approximations: While the mapping to a fully connected Ising model simplifies analysis, it is an approximation. The paper acknowledges that higher-order finite-size corrections are logarithmic and can be neglected for the leading order FSS.
  • Future Work: The authors suggest that the observed two-parameter scaling in annealed networks is a generic feature and warrants further investigation into its underlying mechanisms. They also point out that understanding the distinct FSS behaviors between annealed and quenched networks, particularly the role of quenched disorder and dynamic degrees of freedom, remains a significant challenge. The paper also hints at ongoing investigations into these aspects.

Summary updated

Abstract

We study the critical behavior of the Ising model in annealed scale-free (SF) networks of finite system size with forced upper cutoff in degree. By mapping the model onto the weighted fully connected Ising model, we derive analytic results for the finite-size scaling (FSS) near the phase transition, characterized by the cutoff-dependent two-parameter scaling with four distinct scaling regimes, in highly heterogeneous networks. These results are essentially the same as those found for the nonequilibrium contact process in annealed SF networks, except for an additional complication due to the trivial critical point shift in finite systems. The discrepancy of the FSS theories between annealed and quenched SF networks still remains in the equilibrium Ising model, like some other nonequilibrium models. All of our analytic results are confirmed reasonably well by numerical simulations.
BibTeX
@inproceedings{lee2009critical,
  author = {Sang Hoon Lee and Meesoon Ha and Hawoong Jeong and Jae Dong Noh and Hyunggyu Park},
  title = {{Critical behavior of the Ising model in annealed scale-free networks}},
  booktitle = {Physical},
  year = {2009},
  doi = {10.1103/physreve.80.051127},
}

Loading related content...

Figure