Poisson-Gamma Modeling of Inter-Relational Dependencies in Dynamic Knowledge Graphs

2026-07-02

AI Summary

This work addresses the challenge of link prediction in dynamic knowledge graphs under conditions of data sparsity, high noise, and incompleteness, where modeling temporal dynamics and cross-relational dependencies is particularly difficult. The authors propose PGRE, a novel model that introduces a Poisson–Gamma probabilistic framework to dynamic knowledge graphs for the first time. PGRE models multi-relational temporal links via a Poisson–Bernoulli process, represents entity-factor associations through latent variables governed by Gamma distributions, and captures cross-relational dependencies using a shared latent community structure. Temporal evolution of these latent variables is further modeled via a Gamma Markov process. Experimental results demonstrate that PGRE significantly outperforms existing methods on standard benchmarks, especially in sparse settings, and effectively uncovers meaningful patterns of relational evolution.

Abstract

Dynamic knowledge graphs are ubiquitous in today's AI applications, as we represent molecular structures, social relationships, and language information using these graph models. As knowledge graphs evolve over time and are often noisy and incomplete, modeling their temporal and relational dependencies becomes crucial for downstream tasks. To address these challenges, this paper proposes PGRE (Poisson-Gamma Relational Evolution), a probabilistic model for modeling inter-relational dependencies in dynamic knowledge graphs. PGRE represents multi-relational temporal links via a Poisson-Bernoulli formulation. It introduces Gamma-distributed latent variables to capture entity-factor associations and cross-relation dependencies mediated by shared latent communities. A Gamma Markov process further models the temporal evolution of these latent variables, enabling principled characterization of relational dynamics. Experiments on benchmark datasets show that PGRE achieves competitive performance in link prediction, particularly in sparse settings, while revealing meaningful relational evolution patterns in dynamic knowledge graphs.

Problem

• dynamic knowledge graphs
• inter-relational dependencies
• temporal evolution
• link prediction
• relational dynamics

Innovation

• Poisson-Gamma modeling
• dynamic knowledge graphs
• inter-relational dependencies
• Gamma Markov process
• latent communities