Rotation-equivariant Graph Neural Networks for Learning Glassy Liquids Representations

Abstract

The difficult problem of relating the static structure of glassy liquids and their dynamics is a good target for Machine Learning, an approach which excels at finding complex patterns hidden in data. Indeed, this approach is currently a hot topic in the glassy liquids community, where the state of the art consists in Graph Neural Networks (GNNs), which have great expressive power but are heavy models and lack interpretability. Inspired by recent advances in the field of Machine Learning group-equivariant representations, we build a GNN that learns a robust representation of the glass' static structure by constraining it to preserve the roto-translation (SE(3)) equivariance. We show that this constraint significantly improves the predictive power at comparable or reduced number of parameters but most importantly, improves the ability to generalize to unseen temperatures. While remaining a Deep network, our model has improved interpretability compared to other GNNs, as the action of our basic convolution layer relates directly to well-known rotation-invariant expert features. Through transfer-learning experiments displaying unprecedented performance, we demonstrate that our network learns a robust representation, which allows us to push forward the idea of a learned structural order parameter for glasses.

Submitted to SciPost. 15 pages, 9 figures plus references and 4 pages of appendix

References in corpus (30)

• E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentials
• ChemRL-GEM: Geometry Enhanced Molecular Representation Learning for Property Prediction
• Geometric Deep Learning: Grids, Groups, Graphs, Geodesics, and Gauges
• Identifying structural flow defects in disordered solids using machine learning methods
• Spontaneous and induced dynamic fluctuations in glass-formers I: General results and dependence on ensemble and dynamics
• Predicting plasticity in disordered solids from structural indicators
• Understanding fragility in supercooled Lennard-Jones mixtures. I. Locally preferred structures
• Pre-training Molecular Graph Representation with 3D Geometry
• Autonomously revealing hidden local structures in supercooled liquids
• The Relationship Between Local Structure and Relaxation in Out-of-Equilibrium Glassy Systems
• Structure and dynamics in glass-formers: predictability at large length scales
• Thirty milliseconds in the life of a supercooled liquid
• Locally preferred structures and many-body static correlations in viscous liquids
• EquiformerV2: Improved Equivariant Transformer for Scaling to Higher-Degree Representations
• Averaging local structure to predict the dynamic propensity in supercooled liquids
• Predicting plasticity with soft vibrational modes: from dislocations to glasses
• Assessing the structural heterogeneity of supercooled liquids through community inference
• Equiformer: Equivariant Graph Attention Transformer for 3D Atomistic Graphs
• Predicting dynamic heterogeneity in glass-forming liquids by physics-inspired machine learning
• Elastoplasticity Mediates Dynamical Heterogeneity Below the Mode-Coupling Temperature
• Relevance of Shear Transformations in the Relaxation of Supercooled Liquids
• BOTAN: BOnd TArgeting Network for prediction of slow glassy dynamics by machine learning relative motion
• Dimensionality reduction of local structure in glassy binary mixtures
• Improving the prediction of glassy dynamics by pinpointing the local cage
• Fragility in Glassy Liquids: A Structural Approach Based on Machine Learning
• Towards Multi-spatiotemporal-scale Generalized PDE Modeling
• ForceNet: A Graph Neural Network for Large-Scale Quantum Calculations
• Geometry-enhanced graph neural network for learning the smoothness of glassy dynamics from static structure
• Geometric Clifford Algebra Networks
• Clifford Group Equivariant Neural Networks

Cited by in corpus (5)

• Roadmap on machine learning glassy dynamics
• Selecting Relevant Structural Features for Glassy Dynamics by Information Imbalance
• Graph neural network-based structural classification of glass-forming liquids and its interpretation via self-attention mechanism
• Machine learning that predicts well may not learn the correct physical descriptions of glassy systems
• Interpretability of linear regression models of glassy dynamics