Duality-Induced Regularizer for Tensor Factorization Based Knowledge Graph Completion
This addresses the overfitting problem in knowledge graph completion, which is crucial for applications like recommendation systems and semantic search, though it is incremental as it builds on existing regularization methods.
The paper tackles overfitting in tensor factorization models for knowledge graph completion by proposing a novel regularizer called DURA, which leverages duality between factorization and distance-based models, resulting in consistent and significant performance improvements on benchmarks.
Tensor factorization based models have shown great power in knowledge graph completion (KGC). However, their performance usually suffers from the overfitting problem seriously. This motivates various regularizers -- such as the squared Frobenius norm and tensor nuclear norm regularizers -- while the limited applicability significantly limits their practical usage. To address this challenge, we propose a novel regularizer -- namely, DUality-induced RegulArizer (DURA) -- which is not only effective in improving the performance of existing models but widely applicable to various methods. The major novelty of DURA is based on the observation that, for an existing tensor factorization based KGC model (primal), there is often another distance based KGC model (dual) closely associated with it. Experiments show that DURA yields consistent and significant improvements on benchmarks.