Geometric network comparisons
Dena Marie Asta, Cosma Rohilla Shalizi · Uncertainty in Artificial Intelligence · 2015
Network analysis needs tools to compare networks and assess the significance of differences between networks. We propose a principled statistical approach to network comparison that approximates networks as probability distributions on negatively curved manifolds. We outline the theory, as well as implement the approach on simulated networks, where its accuracy can be confirmed.