Spectral Multigraph Networks for Discovering and Fusing Relationships in Molecules

Abstract

Spectral Graph Convolutional Networks (GCNs) are a generalization of convolutional networks to learning on graph-structured data. Applications of spectral GCNs have been successful, but limited to a few problems where the graph is fixed, such as shape correspondence and node classification. In this work, we address this limitation by revisiting a particular family of spectral graph networks, Chebyshev GCNs, showing its efficacy in solving graph classification tasks with a variable graph structure and size. Chebyshev GCNs restrict graphs to have at most one edge between any pair of nodes. To this end, we propose a novel multigraph network that learns from multi-relational graphs. We model learned edges with abstract meaning and experiment with different ways to fuse the representations extracted from annotated and learned edges, achieving competitive results on a variety of chemical classification benchmarks.

arXiv comment11 pages, 5 figures, NIPS 2018 Workshop on Machine Learning for Molecules and Materials

References in corpus (11)

• Adam: A Method for Stochastic Optimization
• Semi-Supervised Classification with Graph Convolutional Networks
• Inductive Representation Learning on Large Graphs
• Neural Message Passing for Quantum Chemistry
• Spectral Networks and Locally Connected Networks on Graphs
• Striving for Simplicity: The All Convolutional Net
• Deep Convolutional Networks on Graph-Structured Data
• Hierarchical Graph Representation Learning with Differentiable Pooling
• Learning Convolutional Neural Networks for Graphs
• Discriminative Embeddings of Latent Variable Models for Structured Data
• Towards Sparse Hierarchical Graph Classifiers

Cited by in corpus (9)

• Bridging the Gap Between Spectral and Spatial Domains in Graph Neural Networks
• Image Classification with Hierarchical Multigraph Networks
• Revisiting 2D Convolutional Neural Networks for Graph-based Applications
• Mutual Information Maximization in Graph Neural Networks
• Variations on the Chebyshev-Lagrange Activation Function
• Fea2Fea: Exploring Structural Feature Correlations via Graph Neural Networks
• End-to-End Learning from Complex Multigraphs with Latent-Graph Convolutional Networks
• An Experimental Study of the Transferability of Spectral Graph Networks
• DeGNN: Characterizing and Improving Graph Neural Networks with Graph Decomposition