The large deviation principle for interacting dynamical systems on random graphs

DOI10.1007/S00220-022-04312-1MaRDI QIDQ2113491zbMATH OpenOpenAlexFDO

Authors Paul Dupuis, Georgi S. Medvedev

Publication date 14 March 2022

Published in Communications in Mathematical Physics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2007.13899

zbMATH Keywords

interacting particle systemlarge deviationrandom graph

Mathematics Subject Classification ID

Random graphs (graph-theoretic aspects) (05C80) Combinatorial probability (60C05) Large deviations (60F10)

Abstract: Using the weak convergence approach to large deviations, we formulate and prove the large deviation principle (LDP) for W-random graphs in the cut-norm topology. This generalizes the LDP for ErdH{o}s-R{' e}nyi random graphs by Chatterjee and Varadhan. Furthermore, we translate the LDP for random graphs to a class of interacting dynamical systems on such graphs. To this end, we demonstrate that the solutions of the dynamical models depend continuously on the underlying graphs with respect to the cut-norm and apply the contraction principle.

Review text

This paper studies large deviation results for interacting dynamical systems on random graphs. The \(n\) interacting particles follow the evolution \(\dot{u}_i^n=f(u_i^n,\xi_i^n,t)+n^{-1}\sum_{j=1}^nX_{ij}^nD(u_i^n,u_j^n)\), \(u_i^n(0)=g_i^n\), where \(1\le i\le n\), \(u_i^n\) is the state of particle \(i\), \(f\) is the intrinsic dynamics and \(D\) is the pairwise interaction between particles. Here, \(X_{ij}^n\) represents the adjacency matrix of the topology via W-random graphs, the state of each particle \(i\) is generally in \(\mathbb R^d\), \(\{\xi_i^n\}_{i=1}^n\) are random parameters and \(\{g_i^n\}_{i=1}^n\) depict the initial configuration. Some large deviation principles are established for this model and in particular in the setting of W-random graphs. The contraction principle concerning large deviation results is also established for the combinatorial and dynamical systems. Some generalizations and an open problem are provided concerning the main results. The lower semi-continuity of the rate function is also derived.

Recommendations

• A sample-path large deviation principle for dynamic Erdős-Rényi random graphs
• A law of large numbers and large deviations for interacting diffusions on Erdős-Rényi graphs
• The large deviation principle for inhomogeneous Erdős-Rényi random graphs
• scientific article; zbMATH DE number 1191595
• Large deviation principles for empirical measures of colored random graphs

Cites work

• A law of large numbers and large deviations for interacting diffusions on Erdős-Rényi graphs
• A review of the mean field limits for Vlasov equations
• An L^p theory of sparse graph convergence. I: Limits, sparse random graph models, and power law distributions
• An introduction to large deviations for random graphs
• Analysis and approximation of rare events. Representations and weak convergence methods
• Chemical oscillations, waves, and turbulence
• Community detection in sparse networks via Grothendieck's inequality
• scientific article; zbMATH DE number 3629502 (Why is no real title available?)
• scientific article; zbMATH DE number 3353865 (Why is no real title available?)
• Interacting diffusions on random graphs with diverging average degrees: hydrodynamics and large deviations
• Large deviations for random graphs. École d'Été de Probabilités de Saint-Flour XLV -- 2015
• Large networks and graph limits
• Limits of dense graph sequences
• Nonlinear large deviations
• Szemerédi's lemma for the analyst
• The continuum limit of the Kuramoto model on sparse random graphs
• The large deviation principle for the Erdős-Rényi random graph
• The nonlinear heat equation on \(W\)-random graphs
• The nonlinear heat equation on dense graphs and graph limits
• The Vlasov dynamics and its fluctuations in the \(1/N\) limit of interacting classical particles

Cited in

(15)

• Macroscopic determinism in interacting systems using large deviation theory
• Mean-field and graph limits for collective dynamics models with time-varying weights
• Stationarity and uniform in time convergence for the graphon particle system
• Graphon particle system: uniform-in-time concentration bounds
• A sample-path large deviation principle for dynamic Erdős-Rényi random graphs
• Central limit theorems for global and local empirical measures of diffusions on Erdős-Rényi graphs
• Graphon mean field systems
• Large deviations for mean field model in Erdős-Rényi graph
• Stochastic optimization on matrices and a graphon McKean-Vlasov limit
• Non-parametric estimates for graphon mean-field particle systems
• The large deviation principle for W-random spectral measures
• Graph limit for interacting particle systems on weighted random graphs
• Graphon-valued processes with vertex-level fluctuations
• Continuum limit for interacting systems on adaptive networks
• Unified Fourier transform on graphs sampled from stochastic block models

This page was built for publication: The large deviation principle for interacting dynamical systems on random graphs

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2113491)