DeepPath: A Reinforcement Learning Method for Knowledge Graph Reasoning

Abstract

We study the problem of learning to reason in large scale knowledge graphs (KGs). More specifically, we describe a novel reinforcement learning framework for learning multi-hop relational paths: we use a policy-based agent with continuous states based on knowledge graph embeddings, which reasons in a KG vector space by sampling the most promising relation to extend its path. In contrast to prior work, our approach includes a reward function that takes the accuracy, diversity, and efficiency into consideration. Experimentally, we show that our proposed method outperforms a path-ranking based algorithm and knowledge graph embedding methods on Freebase and Never-Ending Language Learning datasets.

EMNLP 17

References in corpus (3)

• Convolutional Neural Networks for Sentence Classification
• Compositional Vector Space Models for Knowledge Base Completion
• Neural Symbolic Machines: Learning Semantic Parsers on Freebase with Weak Supervision

Cited by in corpus (29)

• Deep Reinforcement Learning: An Overview
• UHop: An Unrestricted-Hop Relation Extraction Framework for Knowledge-Based Question Answering
• Multi-Hop Knowledge Graph Reasoning with Reward Shaping
• Interpretable Graph Convolutional Neural Networks for Inference on Noisy Knowledge Graphs
• Iteratively Learning Embeddings and Rules for Knowledge Graph Reasoning
• Variational Knowledge Graph Reasoning
• Knowledge Graph Embedding with Entity Neighbors and Deep Memory Network
• One-Shot Relational Learning for Knowledge Graphs
• Deep Reinforcement Learning for Chinese Zero pronoun Resolution
• Link Prediction using Embedded Knowledge Graphs
• Learning First-Order Rules with Relational Path Contrast for Inductive Relation Reasoning
• Reasoning on Knowledge Graphs with Debate Dynamics
• Learning Collaborative Agents with Rule Guidance for Knowledge Graph Reasoning
• Dynamic Anticipation and Completion for Multi-Hop Reasoning over Sparse Knowledge Graph
• An Overview on the Application of Graph Neural Networks in Wireless Networks
• Collaborative Policy Learning for Open Knowledge Graph Reasoning
• Search from History and Reason for Future: Two-stage Reasoning on Temporal Knowledge Graphs
• Collaborative Graph Walk for Semi-supervised Multi-Label Node Classification
• Recurrent Attention Walk for Semi-supervised Classification
• Reasoning Over Paths via Knowledge Base Completion
• Joint Entity Linking with Deep Reinforcement Learning
• Jointly Learning Explainable Rules for Recommendation with Knowledge Graph
• Recent Advances in Automated Question Answering In Biomedical Domain
• Completion Reasoning Emulation for the Description Logic EL+
• Knowledge Graph Completion with Text-aided Regularization
• Differentiable Representations For Multihop Inference Rules
• EM-RBR: a reinforced framework for knowledge graph completion from reasoning perspective
• Building Rule Hierarchies for Efficient Logical Rule Learning from Knowledge Graphs
• Inferring Substitutable and Complementary Products with Knowledge-Aware Path Reasoning based on Dynamic Policy Network