Saved in:
Bibliographic Details
Main Authors: Li, Rui, Li, Chaozhuo, Shen, Yanming, Zhang, Zeyu, Chen, Xu
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2405.08540
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929343357255680
author Li, Rui
Li, Chaozhuo
Shen, Yanming
Zhang, Zeyu
Chen, Xu
author_facet Li, Rui
Li, Chaozhuo
Shen, Yanming
Zhang, Zeyu
Chen, Xu
contents Recent advances in knowledge graph embedding (KGE) rely on Euclidean/hyperbolic orthogonal relation transformations to model intrinsic logical patterns and topological structures. However, existing approaches are confined to rigid relational orthogonalization with restricted dimension and homogeneous geometry, leading to deficient modeling capability. In this work, we move beyond these approaches in terms of both dimension and geometry by introducing a powerful framework named GoldE, which features a universal orthogonal parameterization based on a generalized form of Householder reflection. Such parameterization can naturally achieve dimensional extension and geometric unification with theoretical guarantees, enabling our framework to simultaneously capture crucial logical patterns and inherent topological heterogeneity of knowledge graphs. Empirically, GoldE achieves state-of-the-art performance on three standard benchmarks. Codes are available at https://github.com/xxrep/GoldE.
format Preprint
id arxiv_https___arxiv_org_abs_2405_08540
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Generalizing Knowledge Graph Embedding with Universal Orthogonal Parameterization
Li, Rui
Li, Chaozhuo
Shen, Yanming
Zhang, Zeyu
Chen, Xu
Machine Learning
Artificial Intelligence
Recent advances in knowledge graph embedding (KGE) rely on Euclidean/hyperbolic orthogonal relation transformations to model intrinsic logical patterns and topological structures. However, existing approaches are confined to rigid relational orthogonalization with restricted dimension and homogeneous geometry, leading to deficient modeling capability. In this work, we move beyond these approaches in terms of both dimension and geometry by introducing a powerful framework named GoldE, which features a universal orthogonal parameterization based on a generalized form of Householder reflection. Such parameterization can naturally achieve dimensional extension and geometric unification with theoretical guarantees, enabling our framework to simultaneously capture crucial logical patterns and inherent topological heterogeneity of knowledge graphs. Empirically, GoldE achieves state-of-the-art performance on three standard benchmarks. Codes are available at https://github.com/xxrep/GoldE.
title Generalizing Knowledge Graph Embedding with Universal Orthogonal Parameterization
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2405.08540