Graph Cross Networks with Vertex Infomax Pooling

Abstract

We propose a novel graph cross network (GXN) to achieve comprehensive feature learning from multiple scales of a graph. Based on trainable hierarchical representations of a graph, GXN enables the interchange of intermediate features across scales to promote information flow. Two key ingredients of GXN include a novel vertex infomax pooling (VIPool), which creates multiscale graphs in a trainable manner, and a novel feature-crossing layer, enabling feature interchange across scales. The proposed VIPool selects the most informative subset of vertices based on the neural estimation of mutual information between vertex features and neighborhood features. The intuition behind is that a vertex is informative when it can maximally reflect its neighboring information. The proposed feature-crossing layer fuses intermediate features between two scales for mutual enhancement by improving information flow and enriching multiscale features at hidden layers. The cross shape of the feature-crossing layer distinguishes GXN from many other multiscale architectures. Experimental results show that the proposed GXN improves the classification accuracy by 2.12% and 1.15% on average for graph classification and vertex classification, respectively. Based on the same network, the proposed VIPool consistently outperforms other graph-pooling methods.

References in corpus (13)

• Distributed Representations of Words and Phrases and their Compositionality
• Inductive Representation Learning on Large Graphs
• Neural Message Passing for Quantum Chemistry
• Learning deep representations by mutual information estimation and maximization
• Variational Graph Auto-Encoders
• Discrete Signal Processing on Graphs: Sampling Theory
• Discriminative Embeddings of Latent Variable Models for Structured Data
• Adaptive Sampling Towards Fast Graph Representation Learning
• LanczosNet: Multi-Scale Deep Graph Convolutional Networks
• HARP: Hierarchical Representation Learning for Networks
• A Fair Comparison of Graph Neural Networks for Graph Classification
• MILE: A Multi-Level Framework for Scalable Graph Embedding
• Learning Topological Representation for Networks via Hierarchical Sampling

Cited by in corpus (3)

• Graph Pooling via Coarsened Graph Infomax
• Maximum Entropy Weighted Independent Set Pooling for Graph Neural Networks
• Diversified Multiscale Graph Learning with Graph Self-Correction