Functional central limit theorem for the subgraph count of the voter model on dynamic random graphs
Pith reviewed 2026-05-23 00:21 UTC · model grok-4.3
The pith
The vector of subgraph counts in the two-opinion voter model on dynamic random graphs converges after centering and scaling to a multidimensional Gaussian process.
A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.
Core claim
In the regime where the number of vertices grows large, under a proper centering and scaling, the joint functional of the vector of subgraph counts converges to a specific multidimensional Gaussian process.
What carries the argument
The functional central limit theorem applied to the vector of subgraph counts under one-way feedback in the voter model on dynamic random graphs.
If this is right
- The subgraph count vector behaves asymptotically like a Gaussian process.
- This convergence holds jointly across multiple subgraph types defined by opinion patterns.
- Fluctuations in opinion-based subgraphs admit approximation by the limiting process when the network is large.
Where Pith is reading between the lines
- The Gaussian process limit may allow efficient large-scale simulation of opinion dynamics without tracking every edge.
- Similar functional limit theorems could be derived for other interacting systems on dynamic graphs with opinion-dependent rewiring.
- The one-way feedback assumption might be relaxed in future work if the reverse influence remains small.
Load-bearing premise
The opinion dynamics remain completely unaffected by the changing graph structure.
What would settle it
Simulations of the model with a large but finite number of vertices in which the scaled subgraph count processes deviate markedly from Gaussian behavior would falsify the claimed convergence.
read the original abstract
In this paper we consider two-opinion voter models on dynamic random graphs, in which the joint dynamics of opinions and graphs acts as one-way feedback, i.e., edges appear and disappear over time depending on the opinions of the two connected vertices, while the opinion dynamics is not affected by the graph structure. Our goal is to investigate the joint evolution of the entries of a voter subgraph count vector, i.e., vector of subgraphs where each vertex has a specific opinion, in the regime that the number of vertices grows large. The main result of this paper is a functional central limit theorem. In particular, we prove that, under a proper centering and scaling, the joint functional of the vector of subgraph counts converges to a specific multidimensional Gaussian process.
Editorial analysis
A structured set of objections, weighed in public.
Referee Report
Summary. The paper considers two-opinion voter models on dynamic random graphs with one-way feedback (edges evolve based on vertex opinions, but opinions evolve independently of the graph). It establishes a functional central limit theorem: under suitable centering and scaling, the joint process of the subgraph-count vector converges to a specific multidimensional Gaussian process as the number of vertices n tends to infinity.
Significance. If the result holds, it supplies a rigorous fluctuation limit for subgraph statistics in a one-way coupled opinion-graph system. This extends classical FCLTs for voter models to dynamic-graph settings and identifies an explicit limiting Gaussian process, which could support further calculations of covariances or long-time behavior. The one-way feedback structure keeps the model tractable while remaining relevant to applications in network opinion dynamics.
major comments (1)
- [Abstract] Abstract: the convergence result is stated without proof details, error bounds, or explicit assumptions beyond the model description, so it is impossible to verify whether the mathematics supports the claim as stated.
Simulated Author's Rebuttal
We thank the referee for reviewing our manuscript. The abstract provides a concise statement of the main result, as is conventional; the full model assumptions, centering/scaling, and proof appear in the body of the paper.
read point-by-point responses
-
Referee: [Abstract] Abstract: the convergence result is stated without proof details, error bounds, or explicit assumptions beyond the model description, so it is impossible to verify whether the mathematics supports the claim as stated.
Authors: Abstracts are high-level summaries by design and do not contain proofs or full technical details. The manuscript defines the dynamic random graph and one-way feedback voter model in Section 2, states the precise assumptions, centering, and scaling in the statement of the main theorem (Theorem 3.1), and provides the complete proof of the functional central limit theorem in Section 4. The limiting multidimensional Gaussian process is identified explicitly. Verification of the mathematics is therefore possible from the body of the paper rather than the abstract. revision: no
Circularity Check
No circularity; standard functional CLT derivation self-contained
full rationale
The paper states a functional central limit theorem for the centered and scaled subgraph-count process of a voter model with explicit one-way feedback on dynamic random graphs in the large-n limit. The abstract and model description present the limit as convergence to an external multidimensional Gaussian process under standard centering/scaling, with no self-definitional equations, fitted inputs renamed as predictions, or load-bearing self-citations that reduce the claim to its own inputs. The one-way coupling and regime are stated directly without importing uniqueness theorems or ansatzes from prior author work. This is a normal non-circular outcome for a convergence theorem in interacting particle systems.
Axiom & Free-Parameter Ledger
axioms (2)
- domain assumption The joint dynamics act as one-way feedback: edges appear and disappear depending on the opinions of the two connected vertices, while opinion dynamics are not affected by the graph structure.
- domain assumption The result holds in the regime where the number of vertices grows large.
Forward citations
Cited by 1 Pith paper
-
Co-evolving vertex and edge dynamics in dense graphs
A co-evolving process of colored vertices and state-dependent edges in dense random graphs converges to a limiting Markov process on colored graphons, with Fisher-Wright diffusion for color densities and a color-depen...
Reference graph
Works this paper leans on
- [1]
- [2]
- [3]
- [4]
- [5]
-
[6]
S. Baldassarri and G. Bet. Asymptotic normality of degree counts in a general preferential attachment model. Markov Processes and Related Fields , 28:577–603, 2022
work page 2022
-
[7]
S. Baldassarri, P. Braunsteins, F. den Hollander, and M. Mandjes. Opinion dynamics on dense dynamic random graphs. arXiv preprint, arXiv:2410.14618 , 2024
-
[8]
R. Basu and A. Sly. Evolving voter model on dense random graphs.The Annals of Applied Probability, 27:1235–1288, 2017
work page 2017
-
[9]
B. B. Bhattacharya, A. Chatterjee, and S. Janson. Fluctuations of subgraph counts in graphon based random graphs. Combinatorics, Probability and Computing , 32:428–464, 2023
work page 2023
-
[10]
Billingsley.Convergence of probability measures
P. Billingsley.Convergence of probability measures. John Wiley & Sons, 2013
work page 2013
-
[11]
Billingsley.Probability and measure
P. Billingsley.Probability and measure. John Wiley & Sons, 2017
work page 2017
-
[12]
M. Bloznelis, J. Karjalainen, and L. Leskelä. Normal and stable approximation to subgraph counts in superpositions of Bernoulli random graphs.Journal of Applied Probability , 61:401–419, 2024
work page 2024
-
[13]
P. Braunsteins, F. den Hollander, and M. Mandjes. Graphon-valued processes with vertex-level fluctuations. arXiv preprint, arXiv:2209.01544 , 2022
- [14]
-
[15]
A. Chatterjee, S. Dan, and B. B. Bhattacharya. Higher-order graphon theory: fluctuations, degeneracies, and inference. arXiv preprint, arXiv:2404.13822 , 2024
-
[16]
S. Chatterjee, D. Sivakoff, and M. Wascher. The effect of avoiding known infected neighbors on the persistence of a recurring infection process.Electronic Journal of Probability , 27:1–40, 2022
work page 2022
- [17]
-
[18]
R. Durrett, J. P. Gleeson, A. L. Lloyd, P. J. Mucha, F. Shi, D. Sivakoff, J. E. Socolar, and C. Varghese. Graph fission in an evolving voter model.Proceedings of the National Academy of Sciences, 109:3682–3687, 2012
work page 2012
-
[19]
P. Erdös and A. Rényi. On the evolution of random graphs.Publication of the Mathematical Institute of the Hungarian Academy of Sciences , 5:17–61, 1960
work page 1960
- [20]
- [21]
- [22]
-
[23]
R. A. Holley and T. M. Liggett. Ergodic theorems for weakly interacting infinite systems and the voter model. The Annals of Probability , 3:643–663, 1975
work page 1975
-
[24]
P. Holme and M. Newman. Nonequilibrium phase transition in the coevolution of networks and opinions. Physical Review E, 74, 2006. 26
work page 2006
-
[25]
E. Jacob and P. Mörters. The contact process on scale-free networks evolving by vertex updating. Royal Society Open Science , 4:1–14, 2017
work page 2017
-
[26]
S. Janson and K. Nowicki. The asymptotic distributions of generalized U-statistics with applications to random graphs.Probability Theory and Related Fields , 90:341–375, 1991
work page 1991
-
[27]
G. Kaur and A. Röllin. Higher-order fluctuations in dense random graph models.Electronic Journal of Probability, 26:1–36, 2021
work page 2021
-
[28]
T. M. Liggett.Interacting Particle Systems . Springer, New York, USA, 1985
work page 1985
-
[29]
K. Nowicki. Asymptotic normality of graph statistics. Journal of Statistical Planning and Inference, 21:209–222, 1989
work page 1989
- [30]
-
[31]
T. Owada. Functional central limit theorem for subgraph counting processes.Electronic Journal of Probability, 22:1–38, 2017
work page 2017
- [32]
- [33]
-
[34]
S. I. Resnick and G. Samorodnitsky. Asymptotic normality of degree counts in a preferential attachment model. Advances in Applied Probability, 48:283–299, 2016
work page 2016
- [35]
-
[36]
P. Sousi and S. Thomas. Cutoff for random walk on dynamical Erdös–Rényi graph.Annales Institut Henri Poincaré, Probability and Statistics , 56:2745–2773, 2020. Mathematical Institute, Leiden University, Einsteinweg 55, 2333 CC Leiden, The Nether- lands Email address: s.baldassarri@math.leidenuniv.nl Korteweg-de Vries Institute for Mathematics, University ...
work page 2020
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.