Conventional representation learning algorithms for knowledge graphs (KG) map each entity to a unique embedding vector, ignoring the rich information contained in the neighborhood. We propose a method named StarGraph, which gives a novel way to utilize the neighborhood information for large-scale knowledge graphs to obtain entity representations. An incomplete two-hop neighborhood subgraph for each target node is at first generated, then processed by a modified self-attention network to obtain the entity representation, which is used to replace the entity embedding in conventional methods. We achieved SOTA performance on ogbl-wikikg2 and got competitive results on fb15k-237. The experimental results proves that StarGraph is efficient in parameters, and the improvement made on ogbl-wikikg2 demonstrates its great effectiveness of representation learning on large-scale knowledge graphs. The code is now available at \url{https://github.com/hzli-ucas/StarGraph}.
@article{arxiv.2205.14209,
title = {StarGraph: Knowledge Representation Learning based on Incomplete Two-hop Subgraph},
author = {Hongzhu Li and Xiangrui Gao and Linhui Feng and Yafeng Deng and Yuhui Yin},
journal= {arXiv preprint arXiv:2205.14209},
year = {2023}
}
Comments
Under review in ICLR 2023 (https://openreview.net/forum?id=mTOB_VK_BWk)