RE: Deep Learning and Knowledge Graphs

Semantic Web Interest Group,

Pascal,



Thank you for the hyperlinks. I enjoyed the publication: Neural-Symbolic Integration and the Semantic Web and observed its hyperlink to: http://www.neural-symbolic.org/ .



Regarding future research and development directions with respect to deep learning and knowledge graphs, we can consider the topics of abstraction [1][2] and conceptual blending [3][4]. We can envision training neural networks to output abstractions from input knowledge graphs and we can envision training neural networks to output conceptual blends from pairs of or sets of input knowledge graphs.





Best regards,

Adam



[1] Saitta, Lorenza, and Jean-Daniel Zucker. Abstraction in artificial intelligence and complex systems. Vol. 456. New York, NY: Springer, 2013.

[2] https://en.wikipedia.org/wiki/Abstraction

[3] Fauconnier, Gilles, and Mark Turner. The way we think: Conceptual blending and the mind's hidden complexities. Basic Books, 2008.

[4] https://en.wikipedia.org/wiki/Conceptual_blending



________________________________
From: Pascal Hitzler <phitzler@googlemail.com>
Sent: Tuesday, May 28, 2019 9:34:39 AM
To: semantic-web@w3.org
Subject: Re: Deep Learning and Knowledge Graphs

Right now you'll find quite a few position papers (by EB members, under
review) at http://www.semantic-web-journal.net/underreview which touch
on this topic.

E.g.

http://www.semantic-web-journal.net/content/neural-symbolic-integration-and-semantic-web

http://www.semantic-web-journal.net/content/role-knowledge-graphs-explainable-ai

http://www.semantic-web-journal.net/content/machine-learning-semantic-web-lessons-learnt-and-next-research-directions

Pascal.

On 5/27/2019 10:35 AM, Adam Sobieski wrote:
> Semantic Web Interest Group,
>
> I’m finding some interesting and introductory publications on the topics
> of artificial neural networks and graph-based data [1][2][3][4]. I’d
> like to share these with the group.
>
> Are there any thoughts on future research and development directions
> with respect to deep learning and knowledge graphs?
>
> Best regards,
>
> Adam Sobieski
>
> [1] Scarselli, Franco, Marco Gori, Ah Chung Tsoi, Markus Hagenbuchner,
> and Gabriele Monfardini. "The graph neural network model." IEEE
> Transactions on Neural Networks 20, no. 1 (2008): 61-80.
>
> [2] Schlichtkrull, Michael, Thomas N. Kipf, Peter Bloem, Rianne van den
> Berg, Ivan Titov, and Max Welling. "Modeling relational data with graph
> convolutional networks." In European Semantic Web Conference, pp.
> 593-607. Springer, Cham, 2018.
>
> [3] Wu, Zonghan, Shirui Pan, Fengwen Chen, Guodong Long, Chengqi Zhang,
> and Philip S. Yu. "A comprehensive survey on graph neural networks."
> arXiv preprint arXiv:1901.00596 (2019).
>
> [4] Zhang, Ziwei, Peng Cui, and Wenwu Zhu. "Deep Learning on Graphs: A
> Survey." arXiv preprint arXiv:1812.04202 (2018).
>

--
Pascal Hitzler
Brage Golding Distinguished Professor of Research
NCR Distinguished Professor   http://www.pascal-hitzler.de
Director of Data Science      http://www.daselab.org
Wright State University       http://www.semantic-web-journal.net/