What Relations Are Reliably Embeddable in Euclidean Space?

Abstract

We consider the problem of embedding a relation, represented as a directed graph, into Euclidean space. For three types of embeddings motivated by the recent literature on knowledge graphs, we obtain characterizations of which relations they are able to capture, as well as bounds on the minimal dimensionality and precision needed.

Cite

Text

Bhattacharjee and Dasgupta. "What Relations Are Reliably Embeddable in Euclidean Space?." Proceedings of the 31st International Conference  on Algorithmic Learning Theory, 2020.

Markdown

[Bhattacharjee and Dasgupta. "What Relations Are Reliably Embeddable in Euclidean Space?." Proceedings of the 31st International Conference  on Algorithmic Learning Theory, 2020.](https://mlanthology.org/alt/2020/bhattacharjee2020alt-relations/)

BibTeX

@inproceedings{bhattacharjee2020alt-relations,
  title     = {{What Relations Are Reliably Embeddable in Euclidean Space?}},
  author    = {Bhattacharjee, Robi and Dasgupta, Sanjoy},
  booktitle = {Proceedings of the 31st International Conference  on Algorithmic Learning Theory},
  year      = {2020},
  pages     = {174-195},
  volume    = {117},
  url       = {https://mlanthology.org/alt/2020/bhattacharjee2020alt-relations/}
}