Stochastic gradient descent methods and uncertainty quantification in extended CLSNA models

2654 

Contributed Abstract 

Paper 

Hancong Pan (1), Xiaojing Zhu (1), Cantay Caliskan (2), Dino Christenson (3), Konstantinos Spiliopoulos (1), Dylan Walker (4), Eric Kolaczyk (5)

(1) Boston University, Boston, MA, USA, (2) University of Rochester, Rochester, NY, USA, (3) Washington University in St. Louis, St. Louis, MO, USA, (4) Chapman University, Orange, CA, USA, (5) McGill University, Montreal, QC, Canada

Xiaojing Zhu  
Boston University

Cantay Caliskan  
University of Rochester

Dino Christenson  
Washington University in St. Louis

Konstantinos Spiliopoulos  
Boston University

Dylan Walker  
Chapman University

Eric Kolaczyk  
McGill University

Hancong Pan  
Boston University

Hancong Pan  
N/A

Coevolving Latent Space Networks with Attractors (CLSNA), introduced by Zhu et al. (2023; JRSS-A), model dynamic networks where nodes in a latent space represent social actors, and edges indicate their interactions. Attractors are added at the latent level to capture the notion of attractive and repulsive forces between nodes, borrowing ideas from dynamical systems theory. The reliance of previous work on MCMC and the requirement for nodes to be present throughout the study period make scaling difficult. We address these issues by (i) introducing an SGD-based parameter estimation method, (ii) developing a novel approach for uncertainty quantification using SGD, and (iii) extending the model to allow nodes to join and leave. Simulation results suggest that our approach results in little loss of accuracy compared to MCMC, but can scale to much larger networks. We revisit Zhu et al.'s analysis of longitudinal social networks for the US Congress from social media X and reinvestigate positive and negative forces among political elites. We now overcome an important selection bias in the previous study and reveal a negative force at play within the Republican Party.

Longitudinal social networks|Attractors|Partisan polarization|Dynamic networks analysis|Co-evolving network model|

International Statistical Institute

Miscellaneous