Graph neural networks for an accurate and interpretable prediction of the properties of polycrystalline materials

Abstract

Various machine learning models have been used to predict the properties of polycrystalline materials, but none of them directly consider the physical interactions among neighboring grains despite such microscopic interactions critically determining macroscopic material properties. Here, we develop a graph neural network (GNN) model for obtaining an embedding of polycrystalline microstructure which incorporates not only the physical features of individual grains but also their interactions. The embedding is then linked to the target property using a feed-forward neural network. Using the magnetostriction of polycrystalline Tb0.3Dy0.7Fe2 alloys as an example, we show that a single GNN model with fixed network architecture and hyperparameters allows for a low prediction error of ~10% over a group of remarkably different microstructures as well as quantifying the importance of each feature in each grain of a microstructure to its magnetostriction. Such a microstructure-graph-based GNN model, therefore, enables an accurate and interpretable prediction of the properties of polycrystalline materials.

Machine learning in energy storage materials

“…Thus, developing ML models with both high predictive ability and high interpretability has been the hot pursuit of materials scientists. [87,113,[204][205][206][207] Recently, physics-informed ML integrates data and mathematical models, which may provide a new insight to discover hidden physics and tackle high-dimensional problems. [50] Altogether, keeping the balance of quantifiable interpretability and intelligent forecasting of ML is a daunting challenge that requires deep integration and collaboration across multiple disciplines.…”

Graph Neural Networks in Particle Physics: Implementations, Innovations, and Challenges

“…In [77], the authors applied layerwise-relevance propagation to characterize the relevant nodes and features for a MLPF model while in [151] the authors apply symbolic regression as a form of disentangled representation learning to extract explicit physical relations including known force laws and Hamiltonians and a new analytic formula that can predict the concentration of dark matter from the mass distribution of nearby cosmic structures. There are also a few examples from the broader physical sciences space; in [152] the authors use an Integrated Gradient method for sensitivity analysis to characterize the relationship between individual material grains and overall material property prediction and in [153] the authors apply a counterfactual perturbation method to understand which components of molecules make them active for disease treatment.…”

Machine learning-assisted design of material properties

“…for molecular design) by quantifying the strength of the contributions of the atom and atom-pair features towards the material property. 145,146 Besides interpreting trained ML models to discover underlying physical laws governing a material property, machine learning techniques can also be used to train accurate yet simple predictive models that are easy to interpret. For example, a recent study 147 reported training neural networks with adjustable parameters quantifying the complexity of the learned functions to find accurate and physically interpretable expressions for predicting a material property of interest.…”