Modeling Relational Data with Graph Convolutional Networks

Abstract

Knowledge graphs enable a wide variety of applications, including question answering and information retrieval. Despite the great effort invested in their creation and maintenance, even the largest (e.g., Yago, DBPedia or Wikidata) remain incomplete. We introduce Relational Graph Convolutional Networks (R-GCNs) and apply them to two standard knowledge base completion tasks: Link prediction (recovery of missing facts, i.e. subject-predicate-object triples) and entity classification (recovery of missing entity attributes). R-GCNs are related to a recent class of neural networks operating on graphs, and are developed specifically to deal with the highly multi-relational data characteristic of realistic knowledge bases. We demonstrate the effectiveness of R-GCNs as a stand-alone model for entity classification. We further show that factorization models for link prediction such as DistMult can be significantly improved by enriching them with an encoder model to accumulate evidence over multiple inference steps in the relational graph, demonstrating a large improvement of 29.8% on FB15k-237 over a decoder-only baseline.

References in corpus (5)

• Inductive Representation Learning on Large Graphs
• Neural Message Passing for Quantum Chemistry
• Variational Graph Auto-Encoders
• Large-scale Simple Question Answering with Memory Networks
• Compositional Vector Space Models for Knowledge Base Completion

Cited by in corpus (47)

• Recent Advances and Applications of Deep Learning Methods in Materials Science
• A Novel Embedding Model for Knowledge Base Completion Based on Convolutional Neural Network
• MolGAN: An implicit generative model for small molecular graphs
• Stochastic Training of Graph Convolutional Networks with Variance Reduction
• Flexible Neural Representation for Physics Prediction
• Adaptive Propagation Graph Convolutional Network
• CoKE: Contextualized Knowledge Graph Embedding
• Relation Embedding with Dihedral Group in Knowledge Graph
• Graph Partition Neural Networks for Semi-Supervised Classification
• KBLRN : End-to-End Learning of Knowledge Base Representations with Latent, Relational, and Numerical Features
• PullNet: Open Domain Question Answering with Iterative Retrieval on Knowledge Bases and Text
• Natural Graph Networks
• Interpretable Graph Convolutional Neural Networks for Inference on Noisy Knowledge Graphs
• Fast Linear Model for Knowledge Graph Embeddings
• Mean-field theory of graph neural networks in graph partitioning
• Graph Neural Networks with Generated Parameters for Relation Extraction
• Link Prediction Adversarial Attack
• Temporal Knowledge Graph Forecasting with Neural ODE
• KELM: Knowledge Enhanced Pre-Trained Language Representations with Message Passing on Hierarchical Relational Graphs
• One-Shot Relational Learning for Knowledge Graphs
• Approximation Ratios of Graph Neural Networks for Combinatorial Problems
• Multi-dimensional Graph Convolutional Networks
• Graph Neural Networks for the Prediction of Substrate-Specific Organic Reaction Conditions
• KECRS: Towards Knowledge-Enriched Conversational Recommendation System
• Open Domain Question Answering Using Early Fusion of Knowledge Bases and Text
• SIRE: Separate Intra- and Inter-sentential Reasoning for Document-level Relation Extraction
• Graph Classification via Deep Learning with Virtual Nodes
• A Re-evaluation of Knowledge Graph Completion Methods
• Attentional Multilabel Learning over Graphs: A Message Passing Approach
• Improving Federated Relational Data Modeling via Basis Alignment and Weight Penalty
• Detecting Hostile Posts using Relational Graph Convolutional Network
• Topological based classification of paper domains using graph convolutional networks
• Learning Graph Pooling and Hybrid Convolutional Operations for Text Representations
• Graph-to-Sequence Learning using Gated Graph Neural Networks
• Learning Predicates as Functions to Enable Few-shot Scene Graph Prediction
• Information Extraction from Scientific Literature for Method Recommendation
• Finding Appropriate Traffic Regulations via Graph Convolutional Networks
• Predicting Semantic Relations using Global Graph Properties
• Spotting Silent Buffer Overflows in Execution Trace through Graph Neural Network Assisted Data Flow Analysis
• Augmenting Compositional Models for Knowledge Base Completion Using Gradient Representations
• Relational dynamic memory networks
• An Experimental Study of the Transferability of Spectral Graph Networks
• Integrating Approaches to Word Representation
• Deep Bayesian Optimization on Attributed Graphs
• Complex network prediction using deep learning
• Semantic Preserving Embeddings for Generalized Graphs
• Anomaly Detection for Aggregated Data Using Multi-Graph Autoencoder