Kronecker Decomposition for Knowledge Graph Embeddings
Knowledge graph embedding research has mainly focused on learning continuous representations of entities and relations tailored towards the link prediction problem. Recent results indicate an ever increasing predictive ability of current approaches on benchmark datasets. However, this effectiveness often comes with the cost of over-parameterization and increased computationally complexity. The former induces extensive hyperparameter optimization to mitigate malicious overfitting. The latter magnifies the importance of winning the hardware lottery. Here, we investigate a remedy for the first problem. We propose a technique based on Kronecker decomposition to reduce the number of parameters in- doi
- 10.1145/3511095.3531276
- name
- Kronecker Decomposition for Knowledge Graph Embeddings
- source
- acm-html-via-r.jina.ai
- acm_url
- https://dl.acm.org/doi/10.1145/3511095.3531276
- authors
- Caglar Demir, Julian Lienen, Axel-Cyrille Ngonga Ngomo
- doi_url
- https://doi.org/10.1145/3511095.3531276
- license
- © 2022 Copyright held by the owner/author(s). Publication rights licensed to ACM.
- summary
- Knowledge graph embedding research has mainly focused on learning continuous representations of entities and relations tailored towards the link prediction problem. Recent results indicate an ever increasing predictive ability of current approaches on benchmark datasets. However, this effectiveness often comes with the cost of over-parameterization and increased computationally complexity. The former induces extensive hyperparameter optimization to mitigate malicious overfitting. The latter magnifies the importance of winning the hardware lottery. Here, we investigate a remedy for the first problem. We propose a technique based on Kronecker decomposition to reduce the number of parameters in
- keywords
- Knowledge Graph Embedding, Kronecker Decomposition, Link Prediction
- source_pdf
- HT-2022_39_31_3511095/3511095.3531276.pdf
- import_kind
- full_text
- open_access
- false
- displayAuthor
- Caglar Demir, Julian Lienen, Axel-Cyrille Ngonga Ngomo
- displayPublishTime
- 2022-06-28
- source_attribution
- Formatting converted from the ACM version of record under supplied ACM publication authorization.
Powered by Seed HypermediaOpen App