HyperKGR: Knowledge Graph Reasoning in Hyperbolic Space with Graph Neural Network Encoding Symbolic Path
Lihui Liu
Abstract
Knowledge graphs (KGs) enable reasoning tasks such as link prediction, question answering, and knowledge discovery.However, real-world KGs are often incomplete, making link prediction both essential and challenging.Existing methods, including embeddingbased and path-based approaches, rely on Euclidean embeddings, which struggle to capture hierarchical structures.GNN-based methods aggregate information through message passing in Euclidean space, but they struggle to effectively encode the recursive treelike structures that emerge in multi-hop reasoning.To address these challenges, instead of learning static entity and relation embeddings, we propose a hyperbolic GNN framework (HYPERKGR) that embeds recursive learning trees in dynamic query-specific hyperbolic space.By incorporating hierarchical message passing, our method naturally aligns with reasoning paths and dynamically adapts to queries, improving prediction accuracy.Unlike static embedding-based approaches, our model learns context-aware embeddings tailored to each query.Experiments on multiple benchmark datasets show that our approach consistently outperforms state-of-the-art methods, demonstrating its effectiveness in KG reasoning.The code can be found in https: //github.com/lihuiliullh/HyperKGR