Predicting missing links between entities in a knowledge graph is a fundamental task to deal with the incompleteness of data on the Web. Knowledge graph embeddings map nodes into a vector space to predict new links, scoring them according to geometric criteria. Relations in the graph may follow patterns that can be learned, e.g., some relations might be symmetric and others might be hierarchical. However, the learning capability of different embedding models varies for each pattern and, so far, no single model can learn all patterns equally well. In this paper, we combine the query representations from several models in a unified one to incorporate patterns that are independently captured by each model. Our combination uses attention to select the most suitable model to answer each query. The models are also mapped onto a non-Euclidean manifold, the Poincar\'e ball, to capture structural patterns, such as hierarchies, besides relational patterns, such as symmetry. We prove that our combination provides a higher expressiveness and inference power than each model on its own. As a result, the combined model can learn relational and structural patterns. We conduct extensive experimental analysis with various link prediction benchmarks showing that the combined model outperforms individual models, including state-of-the-art approaches.

翻译：预测知识图谱中实体间的缺失链接是解决Web数据不完整性的基础任务。知识图谱嵌入通过将节点映射至向量空间，依据几何准则对链接进行评分以预测新链接。图谱中的关系可能遵循可学习的模式，例如某些关系具有对称性，另一些则呈层次结构。然而，不同嵌入模型对各类模式的学习能力存在差异，目前尚无单一模型能同等有效地学习所有模式。本文通过将多个模型的查询表示融合为统一表示，整合各模型独立捕获的模式。该融合方法采用注意力机制选择最适合回答每个查询的模型。同时，将模型映射至非欧几里得流形——庞加莱球，以捕捉层次结构等结构模式及对称性等关系模式。我们证明该融合方法相比各独立模型具有更高的表达力与推理能力，从而使得融合模型能够同时学习关系模式与结构模式。通过使用多种链接预测基准进行广泛实验分析，结果表明融合模型的性能优于包括最先进方法在内的单个模型。