B1087
Title: L-2 regularized maximum likelihood for beta-model estimation in large and sparse networks
Authors: Yuan Zhang - The Ohio State University (United States) [presenting]
Abstract: The beta-model is a powerful tool for modeling networks driven by degree heterogeneity. Its simple yet expressive nature particularly well-suits large and sparse networks, where most models are infeasible due to computational challenge and observation scarcity. However, existing algorithms for beta-model do not scale up; and theoretical understandings remain limited to dense networks. Several major improvements to the method and theory of -model are brought to address the urgent needs of practical applications. The contributions include: 1. Method: we propose a new L-2 penalized MLE scheme; we design a novel algorithm that can comfortably handle sparse networks of millions of nodes, much faster and more memory-parsimonious than any existing algorithm; 2. Theory: we present new error bounds on beta-models under much weaker assumptions; we also establish new lower-bounds and new asymptotic normality results; distinct from existing literature, our results cover both small and large regularization scenarios and reveal their distinct asymptotic dependency structures; 3. Application: we apply our method to large COVID-19 network data sets and discover meaningful results.