Augmenting and Tuning Knowledge Graph Embeddings
This addresses the need for robust and efficient hyperparameter tuning in knowledge graph completion, which is incremental but improves practical usability.
The paper tackles the problem of hyperparameter sensitivity in knowledge graph embeddings for link prediction by proposing an efficient large-scale tuning method using a probabilistic framework and variational expectation-maximization, achieving a new state of the art on standard benchmarks.
Knowledge graph embeddings rank among the most successful methods for link prediction in knowledge graphs, i.e., the task of completing an incomplete collection of relational facts. A downside of these models is their strong sensitivity to model hyperparameters, in particular regularizers, which have to be extensively tuned to reach good performance [Kadlec et al., 2017]. We propose an efficient method for large scale hyperparameter tuning by interpreting these models in a probabilistic framework. After a model augmentation that introduces per-entity hyperparameters, we use a variational expectation-maximization approach to tune thousands of such hyperparameters with minimal additional cost. Our approach is agnostic to details of the model and results in a new state of the art in link prediction on standard benchmark data.