Translating Embeddings for Modeling Multi-relational Data

NeurIPS 2013 Antoine BordesNicolas UsunierAlberto Garcia-DuranJason WestonOksana 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... (read more)

PDF Abstract

Evaluation Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK COMPARE
Link Prediction FB15k TransE MR 125 # 9
Link Prediction FB15k TransE [email protected] 0.471 # 17
Link Prediction WN18RR TransE MRR 0.4659 # 12
Link Prediction WN18RR TransE [email protected] 0.5555 # 11
Link Prediction WN18RR TransE [email protected] 0.4226 # 14