S.N. Dorogovtsev
Abstract:We propose a renormalization group treatment of stochastically growing networks. As an example, we study percolation on growing scale-free networks in the framework of a real-space renormalization group approach. As a result, we find that the critical behavior of percolation on the growing networks differs from that in uncorrelated nets.
What problem does this paper attempt to address?
The problem that this paper attempts to solve is about the percolation problem on **randomly growing networks**, and explores how its critical behavior differs from that of uncorrelated networks. Specifically, the author proposes a **real - space renormalization group approach** to deal with randomly growing networks, and studies the percolation phenomena occurring on these networks through this method.
### Main problems:
1. **The manifestation of the percolation problem in growing networks**: The paper aims to understand the behavior of the percolation problem in randomly growing complex networks, especially how the critical behavior of percolation in growing networks differs from that in static or uncorrelated networks.
2. **Differences in critical behavior**: The author finds that the percolation critical behavior in growing networks is different from that in uncorrelated networks. Specifically:
- **The percolation threshold is zero**: That is, \( p_c = 0 \), which means that even when a small number of edges are retained in the network, there still exists a giant percolating cluster.
- **Singular behavior of the relative size \( M(p) \)**: When \( p \ll 1 \), \( M(p) \sim e^{-\text{const}/p} \), which is different from the behavior in uncorrelated networks.
### Research background:
- **Growth mechanisms of complex networks**: In recent years, complex networks (such as social networks, biological networks, etc.) have attracted extensive research interests. These networks usually have complex connection distributions, especially the **scale - free property**, that is, the degree distribution of nodes follows a power law \( P(k) \propto k^{-\gamma} \).
- **Application of percolation theory**: Percolation theory is an important tool in statistical physics, which is used to study the connectivity of systems under random destruction. In static networks, the percolation problem has been widely studied, but for dynamically growing networks, the percolation behavior is more complex because the structure of the network changes continuously over time.
### Main contributions of the paper:
- **Introduction of the renormalization group method**: The author applies the real - space renormalization group method to randomly growing networks for the first time, showing how to analyze the critical behavior of the network through this method.
- **Revealing new critical behavior**: Through research, the author discovers the unique critical behavior of the percolation problem in growing networks, especially the behavior in the small \( p \) region, which is significantly different from the percolation behavior of traditional uncorrelated networks.
### Summary:
The core problem of this paper is to explore the **percolation critical behavior in randomly growing networks**, and reveals new features of the percolation problem in these networks through the renormalization group method. This research not only deepens the understanding of complex networks, but also provides new ideas for future research on other cooperative models (such as disease spread, Ising model, etc.).