Variational Knowledge Graph Reasoning

Proceedings of NAACL-HLT 2018, pages 1823–1832 New Orleans, Louisiana, June 1 - 6, 2018. c⃝2018 Association for Computational Linguistics Variational Knowledge Graph Reasoning Wenhu Chen, Wenhan Xiong, Xifeng Yan, William Yang Wang Department of Computer Science University of California, Santa Barbara Santa Barbara, CA 93106 {wenhuchen,xwhan,xyan,william}@cs.ucsb.edu Abstract Inferring missing links in knowledge graphs (KG) has attracted a lot of attention from the research community. In this paper, we tackle a practical query answering task involving pre- dicting the relation of a given entity pair. We frame this prediction problem as an inference problem in a probabilistic graphical model and aim at resolving it from a variational infer- ence perspective. In order to model the rela- tion between the query entity pair, we assume that there exists an underlying latent variable (paths connecting two nodes) in the KG, which carries the equivalent semantics of their rela- tions. However, due to the intractability of connections in large KGs, we propose to use variation inference to maximize the evidence lower bound. More specifically, our frame- work (DIVA) is composed of three modules, i.e. a posterior approximator, a prior (path finder), and a likelihood (path reasoner). By using variational inference, we are able to in- corporate them closely into a unified archi- tecture and jointly optimize them to perform KG reasoning. With active interactions among these sub-modules, DIVA is better at handling noise and coping with more complex reason- ing scenarios. In order to evaluate our method, we conduct the experiment of the link pre- diction task on multiple datasets and achieve state-of-the-art performances on both datasets. 1 Introduction Large-scaled knowledge graph supports a lot of downstream natural language processing tasks like question answering, response generation, etc. However, there are large amount of important facts missing in existing KG, which has significantly limited the capability of KG’s application. There- fore, automated reasoning, or the ability for com- puting systems to make new inferences from the observed evidence, has attracted lots of attention from the research community. In recent years, there are surging interests in designing machine learning algorithms for complex reasoning tasks, especially in large knowledge graphs (KGs) where the countless entities and links have posed great challenges to traditional logic-based algorithms. Specifically, we situate our study in this large KG multi-hop reasoning scenario, where the goal is to design an automated inference model to com- plete the missing links between existing entities in large KGs. For examples, if the KG contains a fact like president(BarackObama, USA) and spouse(Michelle, BarackObama), then we would like the machines to complete the missing link livesIn(Michelle, USA) automatically. Systems for this task are essential to complex question answer- ing applications. To tackle the multi-hop link prediction problem, various approaches have been proposed. Some earlier works like PRA (Lao et al., 2011; Gard- ner et al., 2014, 2013) use bounded-depth ran- dom walk with restarts to obtain paths. More re- cently, DeepPath (Xiong et al., 2017) and MIN- ERVA (Das et al., 2018), frame the path-finding problem as a …

Content truncated to 500 words