Path Integral Based Convolution and Pooling for Heterogeneous Graph Neural Networks
This work addresses the challenge of applying graph neural networks to complex heterogeneous datasets, but it is incremental as it builds upon an existing method.
The paper tackles the problem of extending graph neural networks to handle heterogeneous graphs with both node and edge features, by generalizing a path integral-based convolution and pooling method from homogeneous graphs, resulting in a framework that incorporates these features into the model.
Graph neural networks (GNN) extends deep learning to graph-structure dataset. Similar to Convolutional Neural Networks (CNN) using on image prediction, convolutional and pooling layers are the foundation to success for GNN on graph prediction tasks. In the initial PAN paper, it uses a path integral based graph neural networks for graph prediction. Specifically, it uses a convolution operation that involves every path linking the message sender and receiver with learnable weights depending on the path length, which corresponds to the maximal entropy random walk. It further generalizes such convolution operation to a new transition matrix called maximal entropy transition (MET). Because the diagonal entries of the MET matrix is directly related to the subgraph centrality, it provide a trial mechanism for pooling based on centrality score. While the initial PAN paper only considers node features. We further extends its capability to handle complex heterogeneous graph including both node and edge features.