Abstract:We consider two classes of natural stochastic processes on finite unlabeled graphs. These are Euclidean stochastic optimization algorithms on the adjacency matrix of weighted graphs and a modified version of the Metropolis MCMC algorithm on stochastic block models over unweighted graphs. In both cases we show that, as the size of the graph goes to infinity, the random trajectories of the stochastic processes converge to deterministic curves on the space of measure-valued graphons. Measure-valued graphons, introduced by Lovász and Szegedy in \cite{lovasz2010decorated}, are a refinement of the concept of graphons that can distinguish between two infinite exchangeable arrays that give rise to the same graphon limit. We introduce new metrics on this space which provide us with a natural notion of convergence for our limit theorems. This notion is equivalent to the convergence of infinite-exchangeable arrays. Under suitable assumptions and a specified time-scaling, the Metropolis chain admits a diffusion limit as the number of vertices go to infinity. We then demonstrate that, in an appropriately formulated zero-noise limit, the stochastic process of adjacency matrices of this diffusion converges to a deterministic gradient flow curve on the space of graphons introduced in\cite{Oh2023}. A novel feature of this approach is that it provides a precise exponential convergence rate for the Metropolis chain in a certain limiting regime. The connection between a natural Metropolis chain commonly used in exponential random graph models and gradient flows on graphons, to the best of our knowledge, is new in the literature as well.

Variations of dynamic random networks: localization approach

On The Increments Of L(Infinity)-Valued Gaussian Processes

Some Limit Results for the Multi-parameter Gaussian Processes with Stationary Increments

Local weak limit of dynamical inhomogeneous random graphs

Local limit of the random degree constrained process

A local central limit theorem for random walks on expander graphs

Some Limit Theorems for Linear Processes Generated by Symmetrically Exchangeable Random Variables

Variational convergence of exchange-driven stochastic particle systems in the thermodynamic limit

Nonlocal Bounded Variations with Applications

A Unifying Framework for Global Gaussianization: Asymptotic Equivalence of Locally Stationary Processes and Bivariate White Noise

Dynamics of non-classical interval exchanges

Local limit theorem for joint subgraph counts

On the Multivariate Generalized Counting Process and its Time-Changed Variants

Local weak convergence for sparse networks of interacting processes

Random centers of localization for random operators

Path convergence of Markov chains on large graphs

Graph-to-local limit for the nonlocal interaction equation

Large Deviations of a Network of Neurons with Dynamic Sparse Random Connections

On the Theoretical Properties of the Exchange Algorithm

Localizing Volatilities

Trickle-Down in Localization Schemes and Applications