Large Network limit for Latent Variable Process Models with Endogeneously Evolving Interactions among Multiple Types
Abstract
Motivated by the opinion and social dynamics literature, we rigorously study dynamically co-evolving latent space network models, where agent opinions evolve based on a latent state space model influenced by and influencing an underlying dynamically evolving graph determining their interactions. In contrast to earlier work, see [18], agents are allowed to be heterogeneous, i.e., of different types, and their dynamic evolution can capture behaviors like attraction and repulsion between and within groups. In addition, the model encodes bidirectional feedback between the latent process and the underlying interaction network, persistence effects are included, and the interactions are localized, i.e., individual agents do not have global information. We produce weak and empirical limit theorems that encode the asymptotic joint distribution of agent opinion trajectories and their interaction network. In doing so, we describe and exploit the complex dependence structure of the model.
Related
- History-Compatible Energy-Stable Finite Element Schemes for Variable-Density Cahn--Hilliard--Navier--Stokes Flows on Evolving Meshes
- Multi-Variable Conformal Prediction: Optimizing Prediction Sets using Multiple Calibration Variables
- LinearPFN: Amortized Variable Selection for Linear Models with Interactions
- Flexible latent variable models on graphs: Laplace approximated inference for multiview network data
- A General Variable Neighborhood Search for Logistic Network Flow Design with Nonlinear Edge Costs and Transshipment Capacities
BibTeX
@misc{ganguly2026large,
title = {{Large Network limit for Latent Variable Process Models with Endogeneously Evolving Interactions among Multiple Types}},
author = {Ankan Ganguly and Konstantinos Spiliopoulos},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.17369},
eprint = {2610.17369},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.17369},
}
RIS
TY - UNPB
TI - Large Network limit for Latent Variable Process Models with Endogeneously Evolving Interactions among Multiple Types
AU - Ankan Ganguly
AU - Konstantinos Spiliopoulos
PY - 2026
DA - 2026/10/12
PB - arXiv
DO - 10.48550/arxiv.2610.17369
UR - https://arxiv.org/abs/2610.17369
AB - Motivated by the opinion and social dynamics literature, we rigorously study dynamically co-evolving latent space network models, where agent opinions evolve based on a latent state space model influenced by and influencing an underlying dynamically evolving graph determining their interactions. In contrast to earlier work, see [18], agents are allowed to be heterogeneous, i.e., of different types, and their dynamic evolution can capture behaviors like attraction and repulsion between and within groups. In addition, the model encodes bidirectional feedback between the latent process and the underlying interaction network, persistence effects are included, and the interactions are localized, i.e., individual agents do not have global information. We produce weak and empirical limit theorems that encode the asymptotic joint distribution of agent opinion trajectories and their interaction network. In doing so, we describe and exploit the complex dependence structure of the model.
ER -