NENN: Incorporate Node and Edge Features in Graph Neural Networks

EasyChair Preprint 3486

Abstract

Graph neural networks (GNNs) have attracted an increasing attention in recent years. However, most existing state-of-the-art graph learning methods only focus on node features and largely ignore the edge features that contain rich information about graphs. In this paper, we propose a novel model to incorporate Node and Edge features in graph Neural Networks (NENN) based on a hierarchical dual-level attention mechanism. Specifically, the node-level attention layer and edge-level attention layer are alternately stacked to learn the importance of the node based neighbors and edge based neighbors for each node and edge. Leveraging the proposed NENN, the node and edge embeddings can be mutually reinforced. Extensive experiments on academic citation and molecular networks have verified the effectiveness of our proposed graph embedding model.

Keyphrases: Attention Mechanism, Graph Neural Networks, edge features

@booklet{EasyChair:3486,
  author    = {Yulei Yang and Dongsheng Li},
  title     = {NENN: Incorporate Node and Edge Features in Graph Neural Networks},
  howpublished = {EasyChair Preprint 3486},
  year      = {EasyChair, 2020}}