Weakly-supervised Knowledge Graph Alignment with Adversarial Learning
This addresses the challenge of knowledge graph alignment in domains where obtaining aligned data is expensive or unavailable, offering a solution that can work with little to no supervision.
The paper tackles the problem of aligning knowledge graphs from different sources or languages without requiring many aligned triplets, proposing an unsupervised adversarial learning framework with a mutual information regularization term to prevent mode collapse, and shows effectiveness in unsupervised and weakly-supervised settings on multiple datasets.
This paper studies aligning knowledge graphs from different sources or languages. Most existing methods train supervised methods for the alignment, which usually require a large number of aligned knowledge triplets. However, such a large number of aligned knowledge triplets may not be available or are expensive to obtain in many domains. Therefore, in this paper we propose to study aligning knowledge graphs in fully-unsupervised or weakly-supervised fashion, i.e., without or with only a few aligned triplets. We propose an unsupervised framework to align the entity and relation embddings of different knowledge graphs with an adversarial learning framework. Moreover, a regularization term which maximizes the mutual information between the embeddings of different knowledge graphs is used to mitigate the problem of mode collapse when learning the alignment functions. Such a framework can be further seamlessly integrated with existing supervised methods by utilizing a limited number of aligned triples as guidance. Experimental results on multiple datasets prove the effectiveness of our proposed approach in both the unsupervised and the weakly-supervised settings.