Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Bridging Computational Social Science and Deep Learning: Cultural Dissemination-Inspired Graph Neural Networks

Asela Hevapathige

TL;DR

AxelGNN is introduced, a novel architecture based on Axelrod's cultural dissemination model that incorporates three key innovations that adaptively promote convergence or divergence based on feature similarity, and achieves competitive or superior performance compared to existing methods in node classification and influence estimation while maintaining computational efficiency.

Abstract

Graph Neural Networks (GNNs) have become vital in applications like document classification in citation networks, epidemic forecasting, viral marketing, user recommendation in social networks, and network monitoring. However, their deployment faces three key challenges: feature oversmoothing in deep architectures, poor handling of heterogeneous relationships, and monolithic feature aggregation. To address these, we introduce AxelGNN, a novel architecture based on Axelrod's cultural dissemination model that incorporates three key innovations: (1) similarity-gated interactions that adaptively promote convergence or divergence based on feature similarity, (2) segment-wise feature copying that enables fine-grained aggregation of semantic feature groups rather than monolithic vectors, and (3) global polarization that maintains multiple distinct representation clusters to prevent oversmoothing. This model demonstrates empirically the capability to handle both homophilic and heterophilic graphs within a single architecture, without requiring specialized model selection based on graph characteristics. Our experiments demonstrate that AxelGNN achieves competitive or superior performance compared to existing methods in node classification and influence estimation while maintaining computational efficiency.

Bridging Computational Social Science and Deep Learning: Cultural Dissemination-Inspired Graph Neural Networks

TL;DR

AxelGNN is introduced, a novel architecture based on Axelrod's cultural dissemination model that incorporates three key innovations that adaptively promote convergence or divergence based on feature similarity, and achieves competitive or superior performance compared to existing methods in node classification and influence estimation while maintaining computational efficiency.

Abstract

Graph Neural Networks (GNNs) have become vital in applications like document classification in citation networks, epidemic forecasting, viral marketing, user recommendation in social networks, and network monitoring. However, their deployment faces three key challenges: feature oversmoothing in deep architectures, poor handling of heterogeneous relationships, and monolithic feature aggregation. To address these, we introduce AxelGNN, a novel architecture based on Axelrod's cultural dissemination model that incorporates three key innovations: (1) similarity-gated interactions that adaptively promote convergence or divergence based on feature similarity, (2) segment-wise feature copying that enables fine-grained aggregation of semantic feature groups rather than monolithic vectors, and (3) global polarization that maintains multiple distinct representation clusters to prevent oversmoothing. This model demonstrates empirically the capability to handle both homophilic and heterophilic graphs within a single architecture, without requiring specialized model selection based on graph characteristics. Our experiments demonstrate that AxelGNN achieves competitive or superior performance compared to existing methods in node classification and influence estimation while maintaining computational efficiency.

Paper Structure

This paper contains 42 sections, 16 equations, 8 figures, 6 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Traditional GNNs
  4. Heterophily-aware GNNs
  5. Oversmoothing in GNNs
  6. Novelty of Our work
  7. Background and Motivation
  8. Axelrod's Cultural Dissemination Model
  9. Convergence and Polarisation in Axelrod Model
  10. Local Convergence
  11. Global Polarization
  12. Motivation: Addressing Homophily, Heterophily, and Oversmoothing in GNNs
  13. Mapping to Homophily/ Heterophily Paradigms
  14. Global Polarization Prevents Oversmoothing
  15. Methodology
  16. ...and 27 more sections

Figures (8)

  • Figure 1: AxelGNN architecture overview showing the layer-wise processing flow with segment-based aggregation. Orange boxes highlight novel contributions: similarity-gated interaction, bistable convergence dynamics, and segment-wise trait copying.
  • Figure 2: Robustness comparison of AxelGNN variants against oversmoothing across different network depths.
  • Figure 3: Parameter sensitivity analysis of segment size (s) in AxelGNN across Pubmed and Film datasets.
  • Figure 4: Runtime efficiency and classification performance comparison on ogbn-arxiv dataset for node classification.
  • Figure 5: Embedding convergence dynamics of AxelGNN variants.
  • ...and 3 more figures

Theorems & Definitions (2)

  • Definition 3.1: Local Convergence
  • Definition 3.2: Global Polarization