Binarized Embeddings for Fast, Space-Efficient Knowledge Graph Completion
Binarized Embeddings for Fast, Space-Efficient Knowledge Graph Completion
复制标题
DOI：
10.1109/tkde.2021.3075070
复制
发表时间：
2023-01
影响因子：
8.9
通讯作者：
Katsuhiko Hayashi;Koki Kishimoto;M. Shimbo
中科院分区：
文献类型：
--
作者：
Katsuhiko Hayashi;Koki Kishimoto;M. Shimbo
基于知识图的向量嵌入的方法已经被积极追求作为一种有前途的方法来完成知识图。然而，现有的嵌入模型生成存储效率低的表示，特别是当实体和关系的数量，以及实值嵌入向量的维数很大时。我们提出了一个二进制化的CANDECOMP/PARAFAC（CP）分解算法，我们称之为B-CP，其中实值参数被替换为二进制值，以减少模型的大小。此外，一个快速的分数计算技术，开发与逐位操作。我们证明了B-CP是充分表达给定一个足够大的嵌入向量的维数。在多个基准数据集上的实验结果表明，该方法成功地将模型大小降低了一个数量级以上，同时保持与实值CP模型相同的任务性能。
Methods based on vector embeddings of knowledge graphs have been actively pursued as a promising approach to knowledge graph completion. However, existing embedding models generate storage-inefficient representations, particularly when the number of entities and relations, and the dimensionality of the real-valued embedding vectors are large. We present a binarized CANDECOMP/PARAFAC (CP) decomposition algorithm, which we refer to as B-CP, where real-valued parameters are replaced by binary values to reduce model size. Moreover, a fast score computation technique is developed with bitwise operations. We prove that B-CP is fully expressive given a sufficiently large dimensionality of embedding vectors. Experimental results on several benchmark datasets demonstrate that the proposed method successfully reduces model size by more than an order of magnitude while maintaining task performance at the same level as the real-valued CP model.