Learning Multi-Relational Semantics Using Neural-Embedding Models

Real-world entities (e.g., people and places) are often connected via relations, forming multi-relational data. Modeling multi-relational data is important in many research areas, from natural language processing to biological data mining [6]. Prior work on multi-relational learning can be categorized into three categories: (1) statistical relational learning (SRL) [10], such as Markov-logic networks [21], which directly encode multi-relational graphs using probabilistic models; (2) path ranking methods [16, 7], which explicitly explore the large relational feature space of relations with random walk; and (3) embedding-based models, which embed multi-relational knowledge into low-dimensional representations of entities and relations via tensor/matrix factorization [24, 18, 19], Bayesian clustering framework [15, 26], and neural networks [20, 1, 2, 25]. Our work focuses on the study of neural-embedding models, where the representations are learned in a neural network architecture. They have shown to be powerful tools for multi-relational learning and inference due to their high scalability and strong generalization abilities. A number of techniques have been recently proposed to learn entity and relation representations using neural networks [1, 2, 25]. They all represent entities as low-dimensional vectors and represent relations as operators that combine the representations of two entities. The main difference amon