Reasoning Over Knowledge Graphs in Vector Space using Embeddings.

Jure Leskovec, Stanford


Answering complex logical queries on large-scale knowledge graphs is a fundamental yet challenging task. In this I will give an overview of using vector space embeddings for performing logical reasoning in knowledge graphs. First, I will talk about knowledge graph completion method that predicts relations between a pair of entities by: Considering the Relational Context of each entity by capturing the relation types adjacent to the entity and modeled through a novel edge-based message passing scheme; Considering the Relational Paths capturing all paths between the two entities; And, adaptively integrating the Relational Context and Relational Path through a learnable attention mechanism. Second, we will also discuss QUERY2BOX, an embedding-based framework for reasoning over arbitrary queries with and, or and existential operators in massive and incomplete KGs. Our main insight is that queries can be embedded as boxes (i.e., hyper-rectangles), where a set of points inside the box corresponds to a set of answer entities of the query. We show that conjunctions can be naturally represented as intersections of boxes and also prove a negative result that handling disjunctions would require embedding with dimension proportional to the number of KG entities.

The slides are available here.

More details in the paper available here.


Jure Leskovec is an associate professor of Computer Science at Stanford University, the Chief Scientist at Pinterest, and an Investigator at the Chan Zuckerberg Biohub. He was the co-founder of a machine learning startup Kosei, which was later acquired by Pinterest. Leskovec's research area is machine learning and data science for large interconnected systems. Focuses on modeling complex, richly-labeled relational structures, graphs, and networks for systems at all scales, from interactions of proteins in a cell to interactions between humans in a society. Applications include commonsense reasoning, recommender systems, social network analysis, computational social science, and computational biology with an emphasis on drug discovery. This research has won several awards including a Lagrange Prize, Microsoft Research Faculty Fellowship, the Alfred P. Sloan Fellowship, and numerous best paper and test of time awards. It has also been featured in popular press outlets such as the New York Times and the Wall Street Journal. Leskovec received his bachelor's degree in computer science from University of Ljubljana, Slovenia, PhD in machine learning from Carnegie Mellon University and postdoctoral training at Cornell University. You can follow him on Twitter at @jure.