Interpretable and Explainable Machine Learning for Materials Science and Chemistry

Abstract

While the uptake of data-driven approaches for materials science and chemistry is at an exciting, early stage, to realise the true potential of machine learning models for successful scientific discovery, they must have qualities beyond purely predictive power. The predictions and inner workings of models should provide a certain degree of explainability by human experts, permitting the identification of potential model issues or limitations, building trust on model predictions and unveiling unexpected correlations that may lead to scientific insights. In this work, we summarize applications of interpretability and explainability techniques for materials science and chemistry and discuss how these techniques can improve the outcome of scientific studies. We discuss various challenges for interpretable machine learning in materials science and, more broadly, in scientific settings. In particular, we emphasize the risks of inferring causation or reaching generalization by purely interpreting machine learning models and the need of uncertainty estimates for model explanations. Finally, we showcase a number of exciting developments in other fields that could benefit interpretability in material science and chemistry problems.

Under review Accounts of Material Research

References in corpus (7)

• Unsupervised Learning of Disentangled and Interpretable Representations from Sequential Data
• Modelling the dielectric constants of crystals using machine learning
• Data-driven discovery of free-form governing differential equations
• An unsupervised learning approach to solving heat equations on chip based on Auto Encoder and Image Gradient
• Theoretical Impediments to Machine Learning With Seven Sparks from the Causal Revolution
• Discovering Equations that Govern Experimental Materials Stability under Environmental Stress using Scientific Machine Learning
• Exploring causal physical mechanisms via non-gaussian linear models and deep kernel learning: applications for ferroelectric domain structures

Cited by in corpus (12)

• Explainable Artificial Intelligence: A Survey of Needs, Techniques, Applications, and Future Direction
• Advances of Machine Learning in Materials Science: Ideas and Techniques
• High-throughput computational screening of nanoporous materials in targeted applications
• Toward Accurate Interpretable Predictions of Materials Properties within Transformer Language Models
• Multimodal Foundation Models for Material Property Prediction and Discovery
• Artificial Intelligence in Materials Science and Engineering: Current Landscape, Key Challenges, and Future Trajectorie
• Analyzing Atomic Interactions in Molecules as Learned by Neural Networks
• Enhancing Model Learning and Interpretation Using Multiple Molecular Graph Representations for Compound Property and Activity Prediction
• A Normalized Descriptor for Unbiased Screening of Second-Order Nonlinear Optical Materials
• The Predictive Power of Chemical Bonding Analysis in Materials: a Perspective on Optoelectronic Properties
• Self-supervised Representations and Node Embedding Graph Neural Networks for Accurate and Multi-scale Analysis of Materials
• Lost Data in Electron Microscopy