Artificial Intelligence in Materials Science and Engineering: Current Landscape, Key Challenges, and Future Trajectorie

Abstract

Artificial Intelligence is rapidly transforming materials science and engineering, offering powerful tools to navigate complexity, accelerate discovery, and optimize material design in ways previously unattainable. Driven by the accelerating pace of algorithmic advancements and increasing data availability, AI is becoming an essential competency for materials researchers. This review provides a comprehensive and structured overview of the current landscape, synthesizing recent advancements and methodologies for materials scientists seeking to effectively leverage these data-driven techniques. We survey the spectrum of machine learning approaches, from traditional algorithms to advanced deep learning architectures, including CNNs, GNNs, and Transformers, alongside emerging generative AI and probabilistic models such as Gaussian Processes for uncertainty quantification. The review also examines the pivotal role of data in this field, emphasizing how effective representation and featurization strategies, spanning compositional, structural, image-based, and language-inspired approaches, combined with appropriate preprocessing, fundamentally underpin the performance of machine learning models in materials research. Persistent challenges related to data quality, quantity, and standardization, which critically impact model development and application in materials science and engineering, are also addressed.

References in corpus (106)

• Adam: A Method for Stochastic Optimization
• XGBoost: A Scalable Tree Boosting System
• Quantum ESPRESSO: a modular and open-source software project for quantum simulations of materials
• Efficient Estimation of Word Representations in Vector Space
• Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift
• An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale
• A Comprehensive Survey on Graph Neural Networks
• Fundamentals of Recurrent Neural Network (RNN) and Long Short-Term Memory (LSTM) Network
• Geometric deep learning: going beyond Euclidean data
• What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?
• Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
• On representing chemical environments
• An Introduction to Variational Autoencoders
• Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
• SchNet - a deep learning architecture for molecules and materials
• E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentials
• A General-Purpose Machine Learning Framework for Predicting Properties of Inorganic Materials
• Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals
• Recent Advances and Applications of Deep Learning Methods in Materials Science
• B-PINNs: Bayesian Physics-Informed Neural Networks for Forward and Inverse PDE Problems with Noisy Data
• A Universal Graph Deep Learning Interatomic Potential for the Periodic Table
• Big Data of Materials Science - Critical Role of the Descriptor
• Hands-on Bayesian Neural Networks -- a Tutorial for Deep Learning Users
• DScribe: Library of Descriptors for Machine Learning in Materials Science
• Comparing molecules and solids across structural and alchemical space
• Machine Learning Unifies the Modelling of Materials and Molecules
• Atomistic Line Graph Neural Network for Improved Materials Property Predictions
• Integrating Machine Learning and Multiscale Modeling: Perspectives, Challenges, and Opportunities in the Biological, Biomedical, and Behavioral Sciences
• Machine Learning of Molecular Electronic Properties in Chemical Compound Space
• AiiDA: Automated Interactive Infrastructure and Database for Computational Science
• Deep Learning Methods for Reynolds-Averaged Navier-Stokes Simulations of Airfoil Flows
• Prediction of Aerodynamic Flow Fields Using Convolutional Neural Networks
• On-the-fly Closed-loop Autonomous Materials Discovery via Bayesian Active Learning
• AtomNet: A Deep Convolutional Neural Network for Bioactivity Prediction in Structure-based Drug Discovery
• ChatGPT Chemistry Assistant for Text Mining and Prediction of MOF Synthesis
• Opportunities and Challenges for Machine Learning in Materials Science
• Machine learning modeling of superconducting critical temperature
• Predicting materials properties without crystal structure: Deep representation learning from stoichiometry
• ChemBERTa: Large-Scale Self-Supervised Pretraining for Molecular Property Prediction
• Interpretable and Explainable Machine Learning for Materials Science and Chemistry
• Materials Cloud, a platform for open computational science
• Deep Generative Modeling for Mechanistic-based Learning and Design of Metamaterial Systems
• Directional Message Passing for Molecular Graphs
• Extracting Accurate Materials Data from Research Papers with Conversational Language Models and Prompt Engineering
• AiiDA 1.0, a scalable computational infrastructure for automated reproducible workflows and data provenance
• Symbolic Regression in Materials Science
• Generative adversarial networks (GAN) based efficient sampling of chemical space for inverse design of inorganic materials
• Global Attention based Graph Convolutional Neural Networks for Improved Materials Property Prediction
• polyBERT: A chemical language model to enable fully machine-driven ultrafast polymer informatics
• FAIR data enabling new horizons for materials research
• Machine-Learning Interatomic Potentials Enable First-Principles Multiscale Modeling of Lattice Thermal Conductivity in Graphene/Borophene Heterostructures
• Perceiver IO: A General Architecture for Structured Inputs & Outputs
• TransPolymer: a Transformer-based language model for polymer property predictions
• Gaussian Processes and Kernel Methods: A Review on Connections and Equivalences
• Machine learning with force-field inspired descriptors for materials: fast screening and mapping energy landscape
• Machine-learned multi-system surrogate models for materials prediction
• Multipole Graph Neural Operator for Parametric Partial Differential Equations
• SMILES Transformer: Pre-trained Molecular Fingerprint for Low Data Drug Discovery
• Graph Neural Networks for an Accurate and Interpretable Prediction of the Properties of Polycrystalline Materials
• ChemBERTa-2: Towards Chemical Foundation Models
• Meta-modeling game for deriving theoretical-consistent, micro-structural-based traction-separation laws via deep reinforcement learning
• GemNet: Universal Directional Graph Neural Networks for Molecules
• Representing Long-Range Context for Graph Neural Networks with Global Attention
• Data augmentation in microscopic images for material data mining
• Representation Benefits of Deep Feedforward Networks
• Simulation of hyperelastic materials in real-time using Deep Learning
• Data-driven discovery of novel 2D materials by deep generative models
• Deep learning for synthetic microstructure generation in a materials-by-design framework for heterogeneous energetic materials
• e3nn: Euclidean Neural Networks
• Maximizing information from chemical engineering data sets: Applications to machine learning
• OPTIMADE, an API for exchanging materials data
• Advances of Machine Learning in Materials Science: Ideas and Techniques
• Crystal Diffusion Variational Autoencoder for Periodic Material Generation
• The Design Space of E(3)-Equivariant Atom-Centered Interatomic Potentials
• Probabilistic Deep Learning for Real-Time Large Deformation Simulations
• Deep learning generative model for crystal structure prediction
• Equiformer: Equivariant Graph Attention Transformer for 3D Atomistic Graphs
• Three-dimensional microstructure generation using generative adversarial neural networks in the context of continuum micromechanics
• MAgNET: A Graph U-Net Architecture for Mesh-Based Simulations
• Periodic Graph Transformers for Crystal Material Property Prediction
• Toward Accurate Interpretable Predictions of Materials Properties within Transformer Language Models
• Linking Machine Learning with Multiscale Numerics: Data-Driven Discovery of Homogenized Equations
• Fine-Tuned Language Models Generate Stable Inorganic Materials as Text
• Large language models as oracles for instantiating ontologies with domain-specific knowledge
• Gaussian process regression + deep neural network autoencoder for probabilistic surrogate modeling in nonlinear mechanics of solids
• Convolution, aggregation and attention based deep neural networks for accelerating simulations in mechanics
• Predicting Superconducting Transition Temperature through Advanced Machine Learning and Innovative Feature Engineering
• Intelligent multiscale simulation based on process-guided composite database
• LLMatDesign: Autonomous Materials Discovery with Large Language Models
• Fracture Properties of Green Nano Fibrous Network with Random and Aligned Fibre Distribution: A Hierarchical Molecular Dynamics and Peridynamics Approach
• A graph-based probabilistic geometric deep learning framework with online enforcement of physical constraints to predict the criticality of defects in porous materials
• MatterGPT: A Generative Transformer for Multi-Property Inverse Design of Solid-State Materials
• Convolutional State Space Models for Long-Range Spatiotemporal Modeling
• Learning Interpretable Models with Causal Guarantees
• Multi-Stage Variational Auto-Encoders for Coarse-to-Fine Image Generation
• AI-driven materials design: a mini-review
• Representation Debiasing of Generated Data Involving Domain Experts
• Data-efficient and Interpretable Inverse Materials Design using a Disentangled Variational Autoencoder
• Unraveling impacts of polycrystalline microstructures on ionic conductivity of ceramic electrolytes by computational homogenization and machine learning
• Crystal Transformer: Self-learning neural language model for Generative and Tinkering Design of Materials
• Controlling risks of AI in chemical science with agents
• Generative Design of Crystal Structures by Point Cloud Representations and Diffusion Model
• Materials Informatics Transformer: A Language Model for Interpretable Materials Properties Prediction
• Using Large Language Models for OntoClean-based Ontology Refinement
• Materials Transformers Language Models for Generative Materials Design: a benchmark study
• Toward Learning Latent-Variable Representations of Microstructures by Optimizing in Spatial Statistics Space

Cited by in corpus (1)

• An Encoder-Decoder Foundation Chemical Language Model for Generative Polymer Design