Projections of Model Spaces for Latent Graph Inference

2023/03/21 by Borde, Haitz Sáez de Ocáriz, Arroyo, Álvaro, Posner, Ingmar · 1 citation
#FOS: Computer and information sciences #Machine Learning (cs.LG)

paper · doi:10.48550/arxiv.2303.11754

Abstract

Graph Neural Networks leverage the connectivity structure of graphs as an inductive bias. Latent graph inference focuses on learning an adequate graph structure to diffuse information on and improve the downstream performance of the model. In this work we employ stereographic projections of the hyperbolic and spherical model spaces, as well as products of Riemannian manifolds, for the purpose of latent graph inference. Stereographically projected model spaces achieve comparable performance to their non-projected counterparts, while providing theoretical guarantees that avoid divergence of the spaces when the curvature tends to zero. We perform experiments on both homophilic and heterophilic graphs.

Cited by

• Bridging Graph and State-Space Modeling for Intensive Care Unit Length of Stay Prediction

Related

• Latent Graph Inference using Product Manifolds
• AMES: A Differentiable Embedding Space Selection Framework for Latent Graph Inference
• Node-Specific Space Selection via Localized Geometric Hyperbolicity in Graph Neural Networks
• Are Hyperbolic Representations in Graphs Created Equal?
• Neural Snowflakes: Universal Latent Graph Inference via Trainable Latent Geometries