Graph Capsule Convolutional Neural Networks

Abstract

Graph Convolutional Neural Networks (GCNNs) are the most recent exciting advancement in deep learning field and their applications are quickly spreading in multi-cross-domains including bioinformatics, chemoinformatics, social networks, natural language processing and computer vision. In this paper, we expose and tackle some of the basic weaknesses of a GCNN model with a capsule idea presented in \cite{hinton2011transforming} and propose our Graph Capsule Network (GCAPS-CNN) model. In addition, we design our GCAPS-CNN model to solve especially graph classification problem which current GCNN models find challenging. Through extensive experiments, we show that our proposed Graph Capsule Network can significantly outperforms both the existing state-of-art deep learning methods and graph kernels on graph classification benchmark datasets.

arXiv commentAccepted at Joint ICML and IJCAI Workshop on Computational Biology, Stockholm, Sweden, 2018

References in corpus (12)

• Semi-Supervised Classification with Graph Convolutional Networks
• The Emerging Field of Signal Processing on Graphs: Extending High-Dimensional Data Analysis to Networks and Other Irregular Domains
• Neural Message Passing for Quantum Chemistry
• Spectral Networks and Locally Connected Networks on Graphs
• Deep Convolutional Networks on Graph-Structured Data
• Biological network comparison using graphlet degree distribution
• Learning Convolutional Neural Networks for Graphs
• On Valid Optimal Assignment Kernels and Applications to Graph Classification
• The Multiscale Laplacian Graph Kernel
• Covariant Compositional Networks For Learning Graphs
• Learning Graph Representations with Embedding Propagation
• Deriving Neural Architectures from Sequence and Graph Kernels

Cited by in corpus (31)

• Hierarchical Graph Representation Learning with Differentiable Pooling
• How Powerful are Graph Neural Networks?
• DisenHAN: Disentangled Heterogeneous Graph Attention Network for Recommendation
• Graph Neural Networks: Taxonomy, Advances and Trends
• Second-Order Pooling for Graph Neural Networks
• Learning Backtrackless Aligned-Spatial Graph Convolutional Networks for Graph Classification
• Power Normalizations in Fine-grained Image, Few-shot Image and Graph Classification
• Graph Signal Processing and Deep Learning: Convolution, Pooling, and Topology
• Graph Star Net for Generalized Multi-Task Learning
• Hierarchical Taxonomy-Aware and Attentional Graph Capsule RCNNs for Large-Scale Multi-Label Text Classification
• Capsule Neural Networks for Graph Classification using Explicit Tensorial Graph Representations
• NGAT4Rec: Neighbor-Aware Graph Attention Network For Recommendation
• Stability and Generalization of Graph Convolutional Neural Networks
• Learning Universal Graph Neural Network Embeddings With Aid Of Transfer Learning
• Revisiting 2D Convolutional Neural Networks for Graph-based Applications
• Simplicial 2-Complex Convolutional Neural Nets
• Graph Convolutional Neural Networks via Motif-based Attention
• A Hierarchy of Graph Neural Networks Based on Learnable Local Features
• DAGCN: Dual Attention Graph Convolutional Networks
• Interpretable Graph Capsule Networks for Object Recognition
• Geometric Scattering for Graph Data Analysis
• Hierarchical Capsule Prediction Network for Marketing Campaigns Effect
• Quaternion Graph Neural Networks
• Segmented Graph-Bert for Graph Instance Modeling
• Capsule Graph Neural Networks with EM Routing
• Graph Attentional Autoencoder for Anticancer Hyperfood Prediction
• Learning Vertex Convolutional Networks for Graph Classification
• Mutual Information Maximization in Graph Neural Networks
• Interpretable Drug Synergy Prediction with Graph Neural Networks for Human-AI Collaboration in Healthcare
• Graph Analysis and Graph Pooling in the Spatial Domain
• Learning-based Efficient Graph Similarity Computation via Multi-Scale Convolutional Set Matching