Revisiting 2D Convolutional Neural Networks for Graph-based Applications

Abstract

Graph convolutional networks (GCNs) are widely used in graph-based applications such as graph classification and segmentation. However, current GCNs have limitations on implementation such as network architectures due to their irregular inputs. In contrast, convolutional neural networks (CNNs) are capable of extracting rich features from large-scale input data, but they do not support general graph inputs. To bridge the gap between GCNs and CNNs, in this paper we study the problem of how to effectively and efficiently map general graphs to 2D grids that CNNs can be directly applied to, while preserving graph topology as much as possible. We therefore propose two novel graph-to-grid mapping schemes, namely, {\em graph-preserving grid layout (GPGL)} and its extension {\em Hierarchical GPGL (H-GPGL)} for computational efficiency. We formulate the GPGL problem as integer programming and further propose an approximate yet efficient solver based on a penalized Kamada-Kawai method, a well-known optimization algorithm in 2D graph drawing. We propose a novel vertex separation penalty that encourages graph vertices to lay on the grid without any overlap. Along with this image representation, even extra 2D maxpooling layers contribute to the PointNet, a widely applied point-based neural network. We demonstrate the empirical success of GPGL on general graph classification with small graphs and H-GPGL on 3D point cloud segmentation with large graphs, based on 2D CNNs including VGG16, ResNet50 and multi-scale maxout (MSM) CNN.

arXiv commentAccepted by T-PAMI. arXiv admin note: substantial text overlap with arXiv:1909.12383

References in corpus (26)

• Adam: A Method for Stochastic Optimization
• Very Deep Convolutional Networks for Large-Scale Image Recognition
• A Comprehensive Survey on Graph Neural Networks
• Semi-Supervised Classification with Graph Convolutional Networks
• Geometric deep learning: going beyond Euclidean data
• Simplifying Graph Convolutional Networks
• GraphRNN: Generating Realistic Graphs with Deep Auto-regressive Models
• Learning Deep Generative Models of Graphs
• Learning Discrete Structures for Graph Neural Networks
• Linked Dynamic Graph CNN: Learning on Point Cloud via Linking Hierarchical Features
• Anonymous Walk Embeddings
• On Valid Optimal Assignment Kernels and Applications to Graph Classification
• The Multiscale Laplacian Graph Kernel
• Second-Order Pooling for Graph Neural Networks
• Graph Capsule Convolutional Neural Networks
• RGCNN: Regularized Graph CNN for Point Cloud Segmentation
• A Deep Generative Model for Graph Layout
• Wasserstein Weisfeiler-Lehman Graph Kernels
• Subgraph Networks with Application to Structural Feature Space Expansion
• DDGK: Learning Graph Representations for Deep Divergence Graph Kernels
• Learning metrics for persistence-based summaries and applications for graph classification
• Building Deep Networks on Grassmann Manifolds
• A Hierarchical Singular Value Decomposition Algorithm for Low Rank Matrices
• Spectral Multigraph Networks for Discovering and Fusing Relationships in Molecules
• When Work Matters: Transforming Classical Network Structures to Graph CNN
• edGNN: a Simple and Powerful GNN for Directed Labeled Graphs