Learning to Exploit Long-term Relational Dependencies in Knowledge Graphs

Lingbing Guo, Zequn Sun, Wei Hu
Proceedings of the 36th International Conference on Machine Learning, PMLR 97:2505-2514, 2019.

Abstract

We study the problem of knowledge graph (KG) embedding. A widely-established assumption to this problem is that similar entities are likely to have similar relational roles. However, existing related methods derive KG embeddings mainly based on triple-level learning, which lack the capability of capturing long-term relational dependencies of entities. Moreover, triple-level learning is insufficient for the propagation of semantic information among entities, especially for the case of cross-KG embedding. In this paper, we propose recurrent skipping networks (RSNs), which employ a skipping mechanism to bridge the gaps between entities. RSNs integrate recurrent neural networks (RNNs) with residual learning to efficiently capture the long-term relational dependencies within and between KGs. We design an end-to-end framework to support RSNs on different tasks. Our experimental results showed that RSNs outperformed state-of-the-art embedding-based methods for entity alignment and achieved competitive performance for KG completion.

Cite this Paper

BibTeX
@InProceedings{pmlr-v97-guo19c, title = {Learning to Exploit Long-term Relational Dependencies in Knowledge Graphs}, author = {Guo, Lingbing and Sun, Zequn and Hu, Wei}, booktitle = {Proceedings of the 36th International Conference on Machine Learning}, pages = {2505--2514}, year = {2019}, editor = {Chaudhuri, Kamalika and Salakhutdinov, Ruslan}, volume = {97}, series = {Proceedings of Machine Learning Research}, month = {09--15 Jun}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v97/guo19c/guo19c.pdf}, url = {https://proceedings.mlr.press/v97/guo19c.html}, abstract = {We study the problem of knowledge graph (KG) embedding. A widely-established assumption to this problem is that similar entities are likely to have similar relational roles. However, existing related methods derive KG embeddings mainly based on triple-level learning, which lack the capability of capturing long-term relational dependencies of entities. Moreover, triple-level learning is insufficient for the propagation of semantic information among entities, especially for the case of cross-KG embedding. In this paper, we propose recurrent skipping networks (RSNs), which employ a skipping mechanism to bridge the gaps between entities. RSNs integrate recurrent neural networks (RNNs) with residual learning to efficiently capture the long-term relational dependencies within and between KGs. We design an end-to-end framework to support RSNs on different tasks. Our experimental results showed that RSNs outperformed state-of-the-art embedding-based methods for entity alignment and achieved competitive performance for KG completion.} }
Endnote
%0 Conference Paper %T Learning to Exploit Long-term Relational Dependencies in Knowledge Graphs %A Lingbing Guo %A Zequn Sun %A Wei Hu %B Proceedings of the 36th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2019 %E Kamalika Chaudhuri %E Ruslan Salakhutdinov %F pmlr-v97-guo19c %I PMLR %P 2505--2514 %U https://proceedings.mlr.press/v97/guo19c.html %V 97 %X We study the problem of knowledge graph (KG) embedding. A widely-established assumption to this problem is that similar entities are likely to have similar relational roles. However, existing related methods derive KG embeddings mainly based on triple-level learning, which lack the capability of capturing long-term relational dependencies of entities. Moreover, triple-level learning is insufficient for the propagation of semantic information among entities, especially for the case of cross-KG embedding. In this paper, we propose recurrent skipping networks (RSNs), which employ a skipping mechanism to bridge the gaps between entities. RSNs integrate recurrent neural networks (RNNs) with residual learning to efficiently capture the long-term relational dependencies within and between KGs. We design an end-to-end framework to support RSNs on different tasks. Our experimental results showed that RSNs outperformed state-of-the-art embedding-based methods for entity alignment and achieved competitive performance for KG completion.
APA
Guo, L., Sun, Z. & Hu, W.. (2019). Learning to Exploit Long-term Relational Dependencies in Knowledge Graphs. Proceedings of the 36th International Conference on Machine Learning, in Proceedings of Machine Learning Research 97:2505-2514 Available from https://proceedings.mlr.press/v97/guo19c.html.

Related Material