arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2302.01849cs.CL
收藏

参数有效的知识图谱嵌入的不可知表示学习

Entity-Agnostic Representation Learning for Parameter-Efficient Knowledge Graph Embedding

College of Computer Science and Technology, Zhejiang University(浙江大学计算机科学与技术学院);School of Software Technology, Zhejiang University(浙江大学软件学院);Donghai Laboratory(东海实验室);Huawei Technologies Co., Ltd.(华为技术有限公司);School of Informatics, The University of Edinburgh(爱丁堡大学信息学院);Alibaba-Zhejiang University Joint Institute of Frontier Technologies(阿里巴巴-浙江大学前沿技术联合研究院)

Mingyang Chen, Wen Zhang, Zhen Yao, Yushan Zhu, Yang Gao, Jeff Z. Pan, Huajun Chen

摘要:

针对嵌入知识图谱所带来的参数存储效率低下的问题,提出了一种与实体无关的表示学习方法.传统的知识图谱嵌入方法通过向知识图谱中的元素分配一个或多个特定嵌入(即,向量表示)。因此,嵌入参数的数量随着知识图谱的增长而线性增加。在我们提出的模型中,不可知表示学习(EARL),我们只学习一小部分实体的嵌入,并将它们称为保留实体。为了获得完整实体集的嵌入,我们从它们的连接关系、k-最近保留实体和多跳邻居中编码它们的可区分信息。我们学习通用和实体不可知的编码器,用于将可区分的信息转换为实体嵌入。这种方法使我们提出的EARL具有静态的,高效的,和较低的参数计数比传统的知识图谱嵌入方法。实验结果表明,EARL使用较少的参数,性能优于基线的链接预测任务,反映了其参数的效率。

英文摘要:

We propose an entity-agnostic representation learning method for handling the problem of inefficient parameter storage costs brought by embedding knowledge graphs. Conventional knowledge graph embedding methods map elements in a knowledge graph, including entities and relations, into continuous vector spaces by assigning them one or multiple specific embeddings (i.e., vector representations). Thus the number of embedding parameters increases linearly as the growth of knowledge graphs. In our proposed model, Entity-Agnostic Representation Learning (EARL), we only learn the embeddings for a small set of entities and refer to them as reserved entities. To obtain the embeddings for the full set of entities, we encode their distinguishable information from their connected relations, k-nearest reserved entities, and multi-hop neighbors. We learn universal and entity-agnostic encoders for transforming distinguishable information into entity embeddings. This approach allows our proposed EARL to have a static, efficient, and lower parameter count than conventional knowledge graph embedding methods. Experimental results show that EARL uses fewer parameters and performs better on link prediction tasks than baselines, reflecting its parameter efficiency.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑