A Statistical Relational Approach to Learning Distance-based GCNs

Abstract

We consider the problem of learning distance-based Graph Convolutional Networks (GCNs) for relational data. Specifically, we first embed the original graph into the Euclidean space ${\mathbb{R}}^m$ using a relational density estimation technique thereby constructing a secondary Euclidean graph. The graph vertices correspond to the target triples and edges denote the Euclidean distances between the target triples. We emphasize the importance of learning the secondary Euclidean graph and the advantages of employing a distance matrix over the typically used adjacency matrix. Our comprehensive empirical evaluation demonstrates the superiority of our approach over $12$ different GCN models, relational embedding techniques and rule learning techniques.

arXiv comment8 pages, 5 figures, 4 tables; accepted to STARAI workshop

References in corpus (7)

• RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Space
• Graph Matching Networks for Learning the Similarity of Graph Structured Objects
• Hypernetwork Knowledge Graph Embeddings
• TensorLog: A Differentiable Deductive Database
• Convolutional Set Matching for Graph Similarity
• HyperKG: Hyperbolic Knowledge Graph Embeddings for Knowledge Base Completion
• Discriminative Gaifman Models