Embedding Models for Episodic Knowledge Graphs

26 Pages Posted: 23 Jan 2019 First Look: Accepted

See all articles by Yunpu Ma

Yunpu Ma

Siemens AG - Corporate Technology

Volker Tresp

Siemens AG

Erik A. Daxberger

ETH Zürich


In recent years a number of large-scale triple-oriented knowledge graphs have been generated and various models have been proposed to perform learning in those graphs. Most knowledge graphs are static and reflect the world in its current state. In reality, of course, the state of the world is changing: a healthy person becomes diagnosed with a disease and a new president is inaugurated. In this paper, we extend models for static knowledge graphs to temporal knowledge graphs. This enables us to store episodic data and to generalize to new facts (inductive learning). We generalize leading learning models for static knowledge graphs (i.e., Tucker, RESCAL, HolE, ComplEx, DistMult) to temporal knowledge graphs. In particular, we introduce a new tensor model, ConT, with superior generalization performance. The performances of all proposed models are analyzed on two different datasets: the Global Database of Events, Language, and Tone (GDELT) and the database for Integrated Conflict Early Warning System (ICEWS). We argue that temporal knowledge graph embeddings might be models also for cognitive episodic memory (facts we remember and can recollect) and that a semantic memory (current facts we know) can be generated from episodic memory by a marginalization operation. We validate this episodic-to-semantic projection hypothesis with the ICEWS dataset.

Keywords: knowledge graph, temporal knowledge graph, semantic memory, episodic memory, tensor models

Suggested Citation

Ma, Yunpu and Tresp, Volker and Daxberger, Erik A., Embedding Models for Episodic Knowledge Graphs (January 21, 2019). Available at SSRN: https://ssrn.com/abstract=3319790 or http://dx.doi.org/10.2139/ssrn.3319790

Yunpu Ma (Contact Author)

Siemens AG - Corporate Technology

Otto-Hahn-Ring 6

Volker Tresp

Siemens AG ( email )

United States

Erik A. Daxberger

ETH Zürich

Rämistrasse 101
Zürich, 8092

Here is the Coronavirus
related research on SSRN

Paper statistics

Abstract Views