Translating Embeddings for Modeling Multi-relational Data

NeurIPS 2013  ·  Antoine Bordes, Nicolas Usunier, Alberto Garcia-Duran, Jason Weston, Oksana Yakhnenko ·

We consider the problem of embedding entities and relationships of multi-relational data in low-dimensional vector spaces. Our objective is to propose a canonical model which is easy to train, contains a reduced number of parameters and can scale up to very large databases... Hence, we propose, TransE, a method which models relationships by interpreting them as translations operating on the low-dimensional embeddings of the entities. Despite its simplicity, this assumption proves to be powerful since extensive experiments show that TransE significantly outperforms state-of-the-art methods in link prediction on two knowledge bases. Besides, it can be successfully trained on a large scale data set with 1M entities, 25k relationships and more than 17M training samples. read more

PDF Abstract

Datasets


Introduced in the Paper:

FB15k WN18

Used in the Paper:

WN18RR FB122

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Benchmark
Link Prediction FB122 TransE HITS@3 58.9 # 5
Hits@5 64.2 # 5
Hits@10 70.2 # 5
MRR 48.0 # 5
Link Prediction FB15k TransE MR 125 # 10
Hits@10 0.471 # 21
Link Prediction FB15k-237 TransE MRR 0.2904 # 38
Hits@10 .4709 # 39
Hits@1 0.1987 # 33
Link Prediction WN18 TransE Hits@10 0.754 # 28
MR 263 # 9
Link Prediction WN18RR TransE MRR 0.4659 # 27
Hits@10 0.5555 # 24
Hits@1 0.4226 # 28

Methods