Perturbation-Robust Predictive Modeling of Social Effects by Network Subspace Generalized Linear Models

2024-10-02

AI Summary

To address the poor robustness and interpretability of conventional models when analyzing noisy network relational data (e.g., social and biological systems), where strong noise interference and complex structural dependencies pose significant challenges, this paper proposes the Network Subspace Generalized Linear Model (NS-GLM). NS-GLM employs subspace-constrained maximum likelihood estimation to simultaneously preserve statistical interpretability and substantially enhance robustness against network structural perturbations. Theoretically, we establish, for the first time, the asymptotic distribution of the estimator under network perturbations. Methodologically, we develop a unified subspace-constrained modeling framework, validated via both stochastic graph simulations and deep learning–based embeddings. Empirical evaluation on large-scale campus conflict data demonstrates that NS-GLM reduces prediction error by 37% and improves estimation stability by a factor of 2.1 compared to baseline methods.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Reasoning under Uncertainty: Relational Probabilistic Models
• Natural Language Processing: Safety and Robustness

Application Category

• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Social Networks and Social Media: Social media analysis through the lenses of networks
• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods

Abstract

Network-linked data, where multivariate observations are interconnected by a network, are becoming increasingly prevalent in fields such as sociology and biology. These data often exhibit inherent noise and complex relational structures, complicating conventional modeling and statistical inference. Motivated by empirical challenges in analyzing such data sets, this paper introduces a family of network subspace generalized linear models designed for analyzing noisy, network-linked data. We propose a model inference method based on subspace-constrained maximum likelihood, which emphasizes flexibility in capturing network effects and provides a robust inference framework against network perturbations. We establish the asymptotic distributions of the estimators under network perturbations, demonstrating the method's accuracy through extensive simulations involving random network models and deep-learning-based embedding algorithms. The proposed methodology is applied to a comprehensive analysis of a large-scale study on school conflicts, where it identifies significant social effects, offering meaningful and interpretable insights into student behaviors.

Problem

• Modeling noisy network-linked data with complex relational structures
• Providing robust inference against network perturbations
• Identifying significant social effects in large-scale network studies

Innovation

• Network subspace generalized linear models for noisy data
• Subspace-constrained maximum likelihood for robust inference
• Asymptotic distribution analysis under network perturbations