Composing Relationships with Translations

By Alberto García-Durán, Antoine Bordes, Nicolas Usunier. 2015

94 citations recorded by the source corpus.

Performing link prediction in Knowledge Bases (KBs) with embedding-based models, like with the model TransE (Bordes et al., 2013) which represents relationships as translations in the embedding space, have shown promising results in recent years. Most of these works are focused on modeling single relationships and hence do not take full advantage of the graph structure of KBs. In this paper, we propose an extension of TransE that learns to explicitly model composition of relationships via the addition of their corresponding translation vectors. We show empirically that this allows to improve performance for predicting single relationships as well as compositions of pairs of them.

Subfields: Artificial Intelligence, Statistical and Nonlinear Physics

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Statistical and Nonlinear Physics

Citations in the corpus

Modeling Relational Data with Graph Convolutional Networks (2017, methodology): “as recently received considerable attention. We can roughly classify previous work into (1) methods creating auxiliary triples, which are then added to the learning objective of a factorization model [15, 13]; (2) approaches using paths (or walks) as features when predicting edges [24]; or (3) doing”
StarSpace: Embed All The Things! (2017, background): “A series of work has been done in this direction, such as (Bordes et al. 2013) and (Garcia-Duran, Bordes, and Usunier 2015).”
Compositional Learning of Embeddings for Relation Paths in Knowledge Base and Text (2016, methodology): “In contrast, using multi-step relation paths (e.g., husband ( barack , michelle ) ∧ mother ( michelle , sasha ) to train KB embeddings has been proposed very recently (Guu et al., 2015; Garcia-Duran et al., 2015; Lin et al., 2015; Neelakantan et al., 2015).”
ProjE: Embedding Projection for Knowledge Graph Completion (2016, methodology): “RTransE [10] solves the relationship-speciﬁc problem in CVSM by using entity and relationship embeddings learned from TransE.”
Analogical Inference for Multi-relational Embeddings (2017, background): “…models where relations are modeled as translation operators in the embedding space, including TransE (Bordes et al., 2013) and its variants TransH (Wang et al., 2014), TransR (Lin et al., 2015b), TransD (Ji et al., 2015), STransE (Nguyen et al., 2016) and RTransE (Garcia-Duran et al., 2015).”
Knowledge Graph Embedding: A Survey of Approaches and Applications (2017, methodology): “A more limited version of the same approach was simultaneously introduced in [97].”