DiffGCN: Graph Convolutional Networks via Differential Operators and Algebraic Multigrid Pooling

Abstract

Graph Convolutional Networks (GCNs) have shown to be effective in handling unordered data like point clouds and meshes. In this work we propose novel approaches for graph convolution, pooling and unpooling, inspired from finite differences and algebraic multigrid frameworks. We form a parameterized convolution kernel based on discretized differential operators, leveraging the graph mass, gradient and Laplacian. This way, the parameterization does not depend on the graph structure, only on the meaning of the network convolutions as differential operators. To allow hierarchical representations of the input, we propose pooling and unpooling operations that are based on algebraic multigrid methods, which are mainly used to solve partial differential equations on unstructured grids. To motivate and explain our method, we compare it to standard convolutional neural networks, and show their similarities and relations in the case of a regular grid. Our proposed method is demonstrated in various experiments like classification and part-segmentation, achieving on par or better than state of the art results. We also analyze the computational cost of our method compared to other GCNs.

References in corpus (12)

• Adam: A Method for Stochastic Optimization
• Very Deep Convolutional Networks for Large-Scale Image Recognition
• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• Semi-Supervised Classification with Graph Convolutional Networks
• Neural Message Passing for Quantum Chemistry
• Spectral Networks and Locally Connected Networks on Graphs
• Fast Graph Representation Learning with PyTorch Geometric
• MeshCNN: A Network with an Edge
• Generative and Discriminative Voxel Modeling with Convolutional Neural Networks
• Dynamic Graph CNN for Learning on Point Clouds
• Linked Dynamic Graph CNN: Learning on Point Cloud via Linking Hierarchical Features
• LeanConvNets: Low-cost Yet Effective Convolutional Neural Networks

Cited by in corpus (1)

• DeltaConv: Anisotropic Operators for Geometric Deep Learning on Point Clouds