Towards Interpretable Sparse Graph Representation Learning with Laplacian Pooling

Abstract

Recent work in graph neural networks (GNNs) has led to improvements in molecular activity and property prediction tasks. Unfortunately, GNNs often fail to capture the relative importance of interactions between molecular substructures, in part due to the absence of efficient intermediate pooling steps. To address these issues, we propose LaPool (Laplacian Pooling), a novel, data-driven, and interpretable hierarchical graph pooling method that takes into account both node features and graph structure to improve molecular representation. We benchmark LaPool on molecular graph prediction and understanding tasks and show that it outperforms recent GNNs. Interestingly, LaPool also remains competitive on non-molecular tasks. Both quantitative and qualitative assessments are done to demonstrate LaPool's improved interpretability and highlight its potential benefits in drug design. Finally, we demonstrate LaPool's utility for the generation of valid and novel molecules by incorporating it into an adversarial autoencoder.

11 pages, with Appendices

References in corpus (21)

• Adam: A Method for Stochastic Optimization
• The Emerging Field of Signal Processing on Graphs: Extending High-Dimensional Data Analysis to Networks and Other Irregular Domains
• Inductive Representation Learning on Large Graphs
• Neural Message Passing for Quantum Chemistry
• Axiomatic Attribution for Deep Networks
• Molecular Graph Convolutions: Moving Beyond Fingerprints
• Deep Convolutional Networks on Graph-Structured Data
• Junction Tree Variational Autoencoder for Molecular Graph Generation
• Discrete Signal Processing on Graphs: Sampling Theory
• GraphRNN: Generating Realistic Graphs with Deep Auto-regressive Models
• MolGAN: An implicit generative model for small molecular graphs
• From Softmax to Sparsemax: A Sparse Model of Attention and Multi-Label Classification
• Learning Deep Generative Models of Graphs
• Gated Graph Sequence Neural Networks
• How Powerful are Graph Neural Networks?
• Graph Convolutional Policy Network for Goal-Directed Molecular Graph Generation
• Wasserstein Auto-Encoders
• DEFactor: Differentiable Edge Factorization-based Probabilistic Graph Generation
• Signal Representations on Graphs: Tools and Applications
• On Controllable Sparse Alternatives to Softmax
• Graph Convolutional Networks with EigenPooling

Cited by in corpus (7)

• Understanding Pooling in Graph Neural Networks
• Hierarchical Representation Learning in Graph Neural Networks with Node Decimation Pooling
• Path Integral Based Convolution and Pooling for Graph Neural Networks
• Optimal Transport Graph Neural Networks
• Disentangling Interpretable Generative Parameters of Random and Real-World Graphs
• MathNet: Haar-Like Wavelet Multiresolution-Analysis for Graph Representation and Learning
• Learnable Structural Semantic Readout for Graph Classification