Bayes in Multi-Layer Networks

Music City Center 

We present an approach for analyzing multilayer networks to address complex inference challenges in fields like security and neuroscience. We introduce a supervised learning framework that leverages inter- and intra-layer dependencies to predict continuous outcomes. Using low-rank models, this method captures intricate relationships, identifies key nodes and edges, and improves computation speed. Its effectiveness is demonstrated on network security data from a national laboratory, significantly enhancing prediction accuracy.