RatE: Relation-Adaptive Translating Embedding for Knowledge Graph Completion

TL;DR

RatE introduces a relation-adaptive translation function with learnable weights in complex space, enhancing expressiveness and reducing ambiguity in knowledge graph completion, leading to state-of-the-art results.

Contribution

The paper proposes a novel relation-adaptive translation function with learnable weights, improving modeling capacity and ambiguity handling in complex vector space for knowledge graph embedding.

Findings

Achieves state-of-the-art performance on four benchmarks.

Improves expressiveness and ambiguity handling with minimal additional parameters.

Introduces a negative sampling method combining prior knowledge and self-adversarial learning.

Abstract

Many graph embedding approaches have been proposed for knowledge graph completion via link prediction. Among those, translating embedding approaches enjoy the advantages of light-weight structure, high efficiency and great interpretability. Especially when extended to complex vector space, they show the capability in handling various relation patterns including symmetry, antisymmetry, inversion and composition. However, previous translating embedding approaches defined in complex vector space suffer from two main issues: 1) representing and modeling capacities of the model are limited by the translation function with rigorous multiplication of two complex numbers; and 2) embedding ambiguity caused by one-to-many relations is not explicitly alleviated. In this paper, we propose a relation-adaptive translation function built upon a novel weighted product in complex space, where the weights are learnable, relation-specific and independent to embedding size. The translation function only requires eight more scalar parameters each relation, but improves expressive power and alleviates embedding ambiguity problem. Based on the function, we then present our Relation-adaptive translating Embedding (RatE) approach to score each graph triple. Moreover, a novel negative sampling method is proposed to utilize both prior knowledge and self-adversarial learning for effective optimization. Experiments verify RatE achieves state-of-the-art performance on four link prediction benchmarks.