Physical Pooling Functions in Graph Neural Networks for Molecular Property Prediction

Abstract

Graph neural networks (GNNs) are emerging in chemical engineering for the end-to-end learning of physicochemical properties based on molecular graphs. A key element of GNNs is the pooling function which combines atom feature vectors into molecular fingerprints. Most previous works use a standard pooling function to predict a variety of properties. However, unsuitable pooling functions can lead to unphysical GNNs that poorly generalize. We compare and select meaningful GNN pooling methods based on physical knowledge about the learned properties. The impact of physical pooling functions is demonstrated with molecular properties calculated from quantum mechanical computations. We also compare our results to the recent set2set pooling approach. We recommend using sum pooling for the prediction of properties that depend on molecular size and compare pooling functions for properties that are molecular size-independent. Overall, we show that the use of physical pooling functions significantly enhances generalization.

References in corpus (17)

• A Comprehensive Survey on Graph Neural Networks
• Neural Message Passing for Quantum Chemistry
• Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
• SchNet - a deep learning architecture for molecules and materials
• Molecular Graph Convolutions: Moving Beyond Fingerprints
• Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals
• Quantum-Chemical Insights from Deep Tensor Neural Networks
• PhysNet: A Neural Network for Predicting Energies, Forces, Dipole Moments and Partial Charges
• Directional Message Passing for Molecular Graphs
• Machine learning of molecular properties: locality and active learning
• Graph Neural Networks for Temperature-Dependent Activity Coefficient Prediction of Solutes in Ionic Liquids
• Molecular Mechanics-Driven Graph Neural Network with Multiplex Graph for Molecular Structures
• Graph Machine Learning for Design of High-Octane Fuels
• Gibbs-Helmholtz Graph Neural Network: capturing the temperature dependency of activity coefficients at infinite dilution
• Symmetrical Graph Neural Network for Quantum Chemistry, with Dual R/K Space
• Graph neural networks for the prediction of molecular structure-property relationships
• Gated Graph Recursive Neural Networks for Molecular Property Prediction

Cited by in corpus (4)

• Generative AI and Process Systems Engineering: The Next Frontier
• Integrating Chemical Language and Molecular Graph in Multimodal Fused Deep Learning for Drug Property Prediction
• Thermodynamics-Consistent Graph Neural Networks
• Predicting the Temperature Dependence of Surfactant CMCs Using Graph Neural Networks