Search papers, labs, and topics across Lattice.
This paper introduces T-GINEE, a novel statistical regularization framework for multilayer graph representation learning that explicitly models inter-layer dependencies using tensor-based generalized estimating equations. T-GINEE leverages CP tensor decomposition to capture structural dependencies through shared latent factors and models inter-layer correlations via working covariance matrices within a generalized estimating equation framework. Experiments on synthetic and real-world datasets demonstrate T-GINEE's effectiveness in capturing complex relationships within multilayer networks.
Untangling multilayer networks just got easier: T-GINEE uses tensors to explicitly model cross-layer dependencies, outperforming methods that treat layers independently.
Traditional network analysis focuses on single-layer networks, real-world systems often form multilayer networks with multiple relationship types. However, existing methods typically fail to capture complex inter-layer dependencies by treating layers independently or aggregating them. To address this, we propose T-GINEE (Tensor-Based Generalized Multilayer-graph Estimating Equation), a statistical regularization framework combining tensor-based generalized estimating equations with task-specific loss to model cross-network correlations explicitly. Key innovations include: (1) CP tensor decomposition capturing structural dependencies via shared latent factors; (2) a generalized estimating equation framework modeling inter-layer correlations through working covariance matrices; and (3) a flexible link function accommodating characteristics like sparsity. Our theoretical analysis establishes consistency and asymptotic normality under mild conditions. Extensive experiments on synthetic and real-world datasets validate T-GINEE's effectiveness for multilayer network analysis.
