Knowledge Graph Completion

Embedding-based Models

  1. ProjE: Embedding projection for knowledge graph completion
  2. Holographic embeddings of knowledge graphs
  3. Learning entity and relation embeddings for knowledge graph completion
  4. Knowledge graph embedding by translating on hyperplanes
  5. Translating embeddings for modeling multi-relational data
  6. Diachronic embedding for temporal knowledge graph completion
  7. TuckER: Tensor factorization for knowledge graph completion

Relation Path Reasoning

  1. Variational knowledge graph reasoning
  2. Chains of reasoning over entities, relations, and text using recurrent neural networks
  3. Compositional vector space models for knowledge base completion
  4. Incorporating vector space similarity in random walk inference over knowledge bases
  5. Relational retrieval using a combination of path-constrained random walks

Reinforcement Learning-based Path Finding

  1. Collaborative policy learning for open knowledge graph reasoning
  2. M-Walk: Learning to walk over graphs using monte carlo tree search
  3. Multi-hop knowledge graph reasoning with reward shaping
  4. Go for a walk and arrive at the answer: Reasoning over paths in knowledge bases using reinforcement learning
  5. DeepPath: A reinforcement learning method for knowledge graph reasoning

Rule-based Reasoning

  1. Efficient probabilistic logic reasoning with graph neural networks
  2. Differentiable learning of numerical rules in knowledge graphs
  3. Probabilistic logic neural networks for reasoning
  4. Iteratively learning embeddings and rules for knowledge graph reasoning
  5. An embedding-based approach to rule learning in knowledge graphs
  6. Knowledge graph embedding with iterative guidance from soft rules
  7. Differentiable learning of logical rules for knowledge base reasoning
  8. End-to-end differentiable proving
  9. Jointly embedding knowledge graphs and logical rules
  10. AMIE: association rule mining under incomplete evidence in ontological knowledge bases

Meta Relational Learning

  1. Generative adversarial zero-shot relational learning for knowledge graphs
  2. Few-shot knowledge graph completion
  3. Meta relational learning for few-shot link prediction in knowledge graphs
  4. Adapting meta knowledge graph information for multi-hop reasoning over few-shot relations
  5. One-shot relational learning for knowledge graphs

Triple Classification

  1. Triple classification using regions and fine-grained entity typing
  2. Analogical inference for multi-relational embeddings
  3. Holographic embeddings of knowledge graphs
  4. Learning entity and relation embeddings for knowledge graph completion
  5. Knowledge graph embedding by translating on hyperplanes
  6. Reasoning with neural tensor networks for knowledge base completion
Ryan

Ryan

Data Scientist

Leave a Reply