DIVINE: A Generative Adversarial Imitation Learning Framework for Knowledge Graph Reasoning

Ruiping Li, Xiang Cheng

Abstract
Knowledge graphs (KGs) often suffer from sparseness and incompleteness. Knowledge graph reasoning provides a feasible way to address such problems. Recent studies on knowledge graph reasoning have shown that reinforcement learning (RL) based methods can provide state-of-the-art performance. However, existing RL-based methods require numerous trials for path-finding and rely heavily on meticulous reward engineering to fit specific dataset, which is inefficient and laborious to apply to fast-evolving KGs. To this end, in this paper, we present DIVINE, a novel plug-and-play framework based on generative adversarial imitation learning for enhancing existing RL-based methods. DIVINE guides the path-finding process, and learns reasoning policies and reward functions self-adaptively through imitating the demonstrations automatically sampled from KGs. Experimental results on two benchmark datasets show that our framework improves the performance of existing RL-based methods while eliminating extra reward engineering.
Anthology ID:
D19-1266
Volume:
Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)
Month:
November
Year:
2019
Address:
Hong Kong, China
Editors:
Kentaro Inui, Jing Jiang, Vincent Ng, Xiaojun Wan
Venues:
EMNLP | IJCNLP
SIG:
SIGDAT
Publisher:
Association for Computational Linguistics
Note:
Pages:
2642–2651
Language:
URL:
https://aclanthology.org/D19-1266
DOI:
10.18653/v1/D19-1266
Bibkey:
Cite (ACL):
Ruiping Li and Xiang Cheng. 2019. DIVINE: A Generative Adversarial Imitation Learning Framework for Knowledge Graph Reasoning. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 2642–2651, Hong Kong, China. Association for Computational Linguistics.
Cite (Informal):
DIVINE: A Generative Adversarial Imitation Learning Framework for Knowledge Graph Reasoning (Li & Cheng, EMNLP-IJCNLP 2019)
Copy Citation:
PDF:
https://preview.aclanthology.org/improve-issue-templates/D19-1266.pdf
PDF Search