A Scalable Inter-edge Correlation Modeling in CopulaGNN for Link Sign Prediction

2026-01-27

AI Summary

This work addresses the challenge posed by negative edges in signed graphs, which violate the homophily assumption and hinder the effectiveness of conventional graph neural networks. To overcome this limitation, the authors extend the CopulaGNN framework by incorporating a Gaussian copula to model latent statistical dependencies among edges. They further introduce a scalable edge-wise correlation mechanism that represents the correlation matrix as the Gram matrix of edge embeddings, substantially reducing the number of parameters. Additionally, they reformulate the conditional probability distribution to significantly lower inference costs. The proposed method achieves predictive performance on par with state-of-the-art models while exhibiting linear convergence speed, thereby demonstrating its efficiency and scalability.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Reasoning under Uncertainty: Graphical Models

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Link sign prediction on a signed graph is a task to determine whether the relationship represented by an edge is positive or negative. Since the presence of negative edges violates the graph homophily assumption that adjacent nodes are similar, regular graph methods have not been applicable without auxiliary structures to handle them. We aim to directly model the latent statistical dependency among edges with the Gaussian copula and its corresponding correlation matrix, extending CopulaGNN (Ma et al., 2021). However, a naive modeling of edge-edge relations is computationally intractable even for a graph with moderate scale. To address this, we propose to 1) represent the correlation matrix as a Gramian of edge embeddings, significantly reducing the number of parameters, and 2) reformulate the conditional probability distribution to dramatically reduce the inference cost. We theoretically verify scalability of our method by proving its linear convergence. Also, our extensive experiments demonstrate that it achieves significantly faster convergence than baselines, maintaining competitive prediction performance to the state-of-the-art models.

Problem

• link sign prediction
• signed graph
• graph homophily
• edge correlation
• scalability

Innovation

• CopulaGNN
• link sign prediction
• Gaussian copula
• scalable correlation modeling
• Gramian representation