IME: Integrating Multi-curvature Shared and Specific Embedding for Temporal Knowledge Graph Completion
Open Access
- 13 May 2024
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1954-1962
Abstract
Temporal Knowledge Graphs (TKGs) incorporate a temporal dimension, allowing for a precise capture of the evolution of knowledge and reflecting the dynamic nature of the real world. Typically, TKGs contain complex geometric structures, with various geometric structures interwoven. However, existing Temporal Knowledge Graph Completion (TKGC) methods either model TKGs in a single space or neglect the heterogeneity of different curvature spaces, thus constraining their capacity to capture these intricate geometric structures. In this paper, we propose a novel Integrating Multi-curvature shared and specific Embedding (IME) model for TKGC tasks. Concretely, IME models TKGs into multi-curvature spaces, including hyperspherical, hyperbolic, and Euclidean spaces. Subsequently, IME incorporates two key properties, namely space-shared property and space-specific property. The space-shared property facilitates the learning of commonalities across different curvature spaces and alleviates the spatial gap caused by the heterogeneous nature of multi-curvature spaces, while the space-specific property captures characteristic features. Meanwhile, IME proposes an Adjustable Multi-curvature Pooling (AMP) approach to effectively retain important information. Furthermore, IME innovatively designs similarity, difference, and structure loss functions to attain the stated objective. Experimental results clearly demonstrate the superior performance of IME over existing state-of-the-art TKGC models.Keywords
All Related Versions
Funding Information
- National Natural Science Foundation of China (92370102, 62272015, U21B2038)
- National Key R&D Program of China (2021ZD0111902)
- R&D Program of Beijing Municipal Education commission (KZ202210005008)
This publication has 37 references indexed in Scilit:
- Multi-Level Interaction Based Knowledge Graph CompletionIEEE/ACM Transactions on Audio, Speech, and Language Processing, 2023
- Temporal knowledge graph representation learning with local and global evolutionsKnowledge-Based Systems, 2022
- Tucker decomposition-based temporal knowledge graph completionKnowledge-Based Systems, 2022
- Learning the Best Pooling Strategy for Visual Semantic Embedding2021
- DyERNIE: Dynamic Evolution of Riemannian Manifold Embeddings for Temporal Knowledge Graph Completion2020
- Modeling Relational Data with Graph Convolutional Networks2018
- Spherical and Hyperbolic Embeddings of DataIEEE Transactions on Pattern Analysis and Machine Intelligence, 2014
- Semantic search2003
- Bidirectional recurrent neural networksIEEE Transactions on Signal Processing, 1997
- The Expression of a Tensor or a Polyadic as a Sum of ProductsJournal of Mathematics and Physics, 1927