Hypernetwork Knowledge Graph Embeddings

Abstract

Knowledge graphs are graphical representations of large databases of facts, which typically suffer from incompleteness. Inferring missing relations (links) between entities (nodes) is the task of link prediction. A recent state-of-the-art approach to link prediction, ConvE, implements a convolutional neural network to extract features from concatenated subject and relation vectors. Whilst results are impressive, the method is unintuitive and poorly understood. We propose a hypernetwork architecture that generates simplified relation-specific convolutional filters that (i) outperforms ConvE and all previous approaches across standard datasets; and (ii) can be framed as tensor factorization and thus set within a well established family of factorization models for link prediction. We thus demonstrate that convolution simply offers a convenient computational means of introducing sparsity and parameter tying to find an effective trade-off between non-linear expressiveness and the number of parameters to learn.

References in corpus (7)

• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• Complex Embeddings for Simple Link Prediction
• RotatE: Knowledge Graph Embedding by Relational Rotation in Complex Space
• Go for a Walk and Arrive at the Answer: Reasoning Over Paths in Knowledge Bases using Reinforcement Learning
• SimplE Embedding for Link Prediction in Knowledge Graphs
• Analogical Inference for Multi-Relational Embeddings
• M-Walk: Learning to Walk over Graphs using Monte Carlo Tree Search

Cited by in corpus (24)

• A Survey on Knowledge Graphs: Representation, Acquisition and Applications
• Knowledge Graphs
• Composition-based Multi-Relational Graph Convolutional Networks
• A Brief Review of Hypernetworks in Deep Learning
• A Review of Knowledge Graph Completion
• CoKE: Contextualized Knowledge Graph Embedding
• Representation Learning for Dynamic Graphs: A Survey
• Prototypical Representation Learning for Relation Extraction
• Principled Weight Initialization for Hypernetworks
• Adaptive Filters in Graph Convolutional Neural Networks
• Cognitive Knowledge Graph Reasoning for One-shot Relational Learning
• Neural Graph Embedding Methods for Natural Language Processing
• QuatDE: Dynamic Quaternion Embedding for Knowledge Graph Completion
• On the Ambiguity of Rank-Based Evaluation of Entity Alignment or Link Prediction Methods
• Knowledge Graph Embedding with Atrous Convolution and Residual Learning
• Efficient Relation-aware Scoring Function Search for Knowledge Graph Embedding
• Type-augmented Relation Prediction in Knowledge Graphs
• Out-of-Vocabulary Entities in Link Prediction
• Searching to Sparsify Tensor Decomposition for N-ary Relational Data
• Convolutional Hypercomplex Embeddings for Link Prediction
• A Statistical Relational Approach to Learning Distance-based GCNs
• FC2T2: The Fast Continuous Convolutional Taylor Transform with Applications in Vision and Graphics
• Learning semantic Image attributes using Image recognition and knowledge graph embeddings
• Scalable knowledge base completion with superposition memories