Graph Reasoning Networks
This work addresses reasoning limitations in graph-based machine learning, but it appears incremental as it builds upon existing GNN methods.
The authors tackled the limited high-level reasoning abilities of graph neural networks by introducing Graph Reasoning Networks, which combine fixed and learned graph representations with a differentiable satisfiability solver, showing comparable performance on real-world datasets and potential on synthetic ones.
Graph neural networks (GNNs) are the predominant approach for graph-based machine learning. While neural networks have shown great performance at learning useful representations, they are often criticized for their limited high-level reasoning abilities. In this work, we present Graph Reasoning Networks (GRNs), a novel approach to combine the strengths of fixed and learned graph representations and a reasoning module based on a differentiable satisfiability solver. While results on real-world datasets show comparable performance to GNN, experiments on synthetic datasets demonstrate the potential of the newly proposed method.