Graph Neural Networks Meet Neural-Symbolic Computing: A Survey and Perspective

  • 2020-05-21 17:11:36
  • Luis C. Lamb, Artur Garcez, Marco Gori, Marcelo Prates, Pedro Avelar, Moshe Vardi
  • 0

Abstract

Neural-symbolic computing has now become the subject of interest of bothacademic and industry research laboratories. Graph Neural Networks (GNN) havebeen widely used in relational and symbolic domains, with widespreadapplication of GNNs in combinatorial optimization, constraint satisfaction,relational reasoning and other scientific domains. The need for improvedexplainability, interpretability and trust of AI systems in general demandsprincipled methodologies, as suggested by neural-symbolic computing. In thispaper, we review the state-of-the-art on the use of GNNs as a model ofneural-symbolic computing. This includes the application of GNNs in severaldomains as well as its relationship to current developments in neural-symboliccomputing.

 

Quick Read (beta)

loading the full paper ...