Defeats GAN: A Simpler Model Outperforms in Knowledge Representation Learning
This work addresses the need for more efficient and effective models in knowledge graph applications, though it appears incremental as it builds on existing embedding methods.
The authors tackled the problem of improving knowledge representation learning by proposing Trans-DLR, a method using dynamic learning rate control and a new negative sampling trick, which outperformed a recent GAN-based approach in experiments.
The goal of knowledge representation learning is to embed entities and relations into a low-dimensional, continuous vector space. How to push a model to its limit and obtain better results is of great significance in knowledge graph's applications. We propose a simple and elegant method, Trans-DLR, whose main idea is dynamic learning rate control during training. Our method achieves remarkable improvement, compared with recent GAN-based method. Moreover, we introduce a new negative sampling trick which corrupts not only entities, but also relations, in different probabilities. We also develop an efficient way, which fully utilizes multiprocessing and parallel computing, to speed up evaluation of the model in link prediction tasks. Experiments show that our method is effective.