InGram: Inductive Knowledge Graph Embedding via Relation Graphs

Jaejun Lee, Chanyoung Chung, Joyce Jiyoung Whang
Proceedings of the 40th International Conference on Machine Learning, PMLR 202:18796-18809, 2023.

Abstract

Inductive knowledge graph completion has been considered as the task of predicting missing triplets between new entities that are not observed during training. While most inductive knowledge graph completion methods assume that all entities can be new, they do not allow new relations to appear at inference time. This restriction prohibits the existing methods from appropriately handling real-world knowledge graphs where new entities accompany new relations. In this paper, we propose an INductive knowledge GRAph eMbedding method, InGram, that can generate embeddings of new relations as well as new entities at inference time. Given a knowledge graph, we define a relation graph as a weighted graph consisting of relations and the affinity weights between them. Based on the relation graph and the original knowledge graph, InGram learns how to aggregate neighboring embeddings to generate relation and entity embeddings using an attention mechanism. Experimental results show that InGram outperforms 14 different state-of-the-art methods on varied inductive learning scenarios.

Cite this Paper

BibTeX
@InProceedings{pmlr-v202-lee23c, title = {{I}n{G}ram: Inductive Knowledge Graph Embedding via Relation Graphs}, author = {Lee, Jaejun and Chung, Chanyoung and Whang, Joyce Jiyoung}, booktitle = {Proceedings of the 40th International Conference on Machine Learning}, pages = {18796--18809}, year = {2023}, editor = {Krause, Andreas and Brunskill, Emma and Cho, Kyunghyun and Engelhardt, Barbara and Sabato, Sivan and Scarlett, Jonathan}, volume = {202}, series = {Proceedings of Machine Learning Research}, month = {23--29 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v202/lee23c/lee23c.pdf}, url = {https://proceedings.mlr.press/v202/lee23c.html}, abstract = {Inductive knowledge graph completion has been considered as the task of predicting missing triplets between new entities that are not observed during training. While most inductive knowledge graph completion methods assume that all entities can be new, they do not allow new relations to appear at inference time. This restriction prohibits the existing methods from appropriately handling real-world knowledge graphs where new entities accompany new relations. In this paper, we propose an INductive knowledge GRAph eMbedding method, InGram, that can generate embeddings of new relations as well as new entities at inference time. Given a knowledge graph, we define a relation graph as a weighted graph consisting of relations and the affinity weights between them. Based on the relation graph and the original knowledge graph, InGram learns how to aggregate neighboring embeddings to generate relation and entity embeddings using an attention mechanism. Experimental results show that InGram outperforms 14 different state-of-the-art methods on varied inductive learning scenarios.} }
Endnote
%0 Conference Paper %T InGram: Inductive Knowledge Graph Embedding via Relation Graphs %A Jaejun Lee %A Chanyoung Chung %A Joyce Jiyoung Whang %B Proceedings of the 40th International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2023 %E Andreas Krause %E Emma Brunskill %E Kyunghyun Cho %E Barbara Engelhardt %E Sivan Sabato %E Jonathan Scarlett %F pmlr-v202-lee23c %I PMLR %P 18796--18809 %U https://proceedings.mlr.press/v202/lee23c.html %V 202 %X Inductive knowledge graph completion has been considered as the task of predicting missing triplets between new entities that are not observed during training. While most inductive knowledge graph completion methods assume that all entities can be new, they do not allow new relations to appear at inference time. This restriction prohibits the existing methods from appropriately handling real-world knowledge graphs where new entities accompany new relations. In this paper, we propose an INductive knowledge GRAph eMbedding method, InGram, that can generate embeddings of new relations as well as new entities at inference time. Given a knowledge graph, we define a relation graph as a weighted graph consisting of relations and the affinity weights between them. Based on the relation graph and the original knowledge graph, InGram learns how to aggregate neighboring embeddings to generate relation and entity embeddings using an attention mechanism. Experimental results show that InGram outperforms 14 different state-of-the-art methods on varied inductive learning scenarios.
APA
Lee, J., Chung, C. & Whang, J.J.. (2023). InGram: Inductive Knowledge Graph Embedding via Relation Graphs. Proceedings of the 40th International Conference on Machine Learning, in Proceedings of Machine Learning Research 202:18796-18809 Available from https://proceedings.mlr.press/v202/lee23c.html.

Related Material