Key points are not available for this paper at this time.
Neural-symbolic computing has now become the subject of interest of both and industry research laboratories. Graph Neural Networks (GNN) have widely used in relational and symbolic domains, with widespread of GNNs in combinatorial optimization, constraint satisfaction, reasoning and other scientific domains. The need for improved, interpretability and trust of AI systems in general demands methodologies, as suggested by neural-symbolic computing. In this, we review the state-of-the-art on the use of GNNs as a model of-symbolic computing. This includes the application of GNNs in several as well as its relationship to current developments in neural-symbolic.
Lamb et al. (Sat,) studied this question.