Molecule Property Prediction and Classification with Graph Hypernetworks

Abstract

Graph neural networks are currently leading the performance charts in learning-based molecule property prediction and classification. Computational chemistry has, therefore, become the a prominent testbed for generic graph neural networks, as well as for specialized message passing methods. In this work, we demonstrate that the replacement of the underlying networks with hypernetworks leads to a boost in performance, obtaining state of the art results in various benchmarks. A major difficulty in the application of hypernetworks is their lack of stability. We tackle this by combining the current message and the first message. A recent work has tackled the training instability of hypernetworks in the context of error correcting codes, by replacing the activation function of the message passing network with a low-order Taylor approximation of it. We demonstrate that our generic solution can replace this domain-specific solution.

References in corpus (11)

• Semi-Supervised Classification with Graph Convolutional Networks
• Neural Message Passing for Quantum Chemistry
• Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
• Molecular Graph Convolutions: Moving Beyond Fingerprints
• Quantum-Chemical Insights from Deep Tensor Neural Networks
• Interaction Networks for Learning about Objects, Relations and Physics
• Invariant and Equivariant Graph Networks
• Covariant Compositional Networks For Learning Graphs
• Neural Message Passing with Edge Updates for Predicting Properties of Molecules and Materials
• Relational Pooling for Graph Representations
• Hyper-Graph-Network Decoders for Block Codes

Cited by in corpus (3)

• An Introduction to Electrocatalyst Design using Machine Learning for Renewable Energy Storage
• From Local Structures to Size Generalization in Graph Neural Networks
• PACIA: Parameter-Efficient Adapter for Few-Shot Molecular Property Prediction