Advances of Machine Learning in Materials Science: Ideas and Techniques

Abstract

In this big data era, the use of large dataset in conjunction with machine learning (ML) has been increasingly popular in both industry and academia. In recent times, the field of materials science is also undergoing a big data revolution, with large database and repositories appearing everywhere. Traditionally, materials science is a trial-and-error field, in both the computational and experimental departments. With the advent of machine learning-based techniques, there has been a paradigm shift: materials can now be screened quickly using ML models and even generated based on materials with similar properties; ML has also quietly infiltrated many sub-disciplinary under materials science. However, ML remains relatively new to the field and is expanding its wing quickly. There are a plethora of readily-available big data architectures and abundance of ML models and software; The call to integrate all these elements in a comprehensive research procedure is becoming an important direction of material science research. In this review, we attempt to provide an introduction and reference of ML to materials scientists, covering as much as possible the commonly used methods and applications, and discussing the future possibilities.

arXiv comment60 pages; 22 figures

References in corpus (124)

• Efficient Estimation of Word Representations in Vector Space
• Model-Agnostic Meta-Learning for Fast Adaptation of Deep Networks
• Quantum Machine Learning
• HuggingFace's Transformers: State-of-the-art Natural Language Processing
• Axiomatic Attribution for Deep Networks
• Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties
• On representing chemical environments
• Striving for Simplicity: The All Convolutional Net
• Supervised learning with quantum enhanced feature spaces
• Machine learning and the physical sciences
• Fast and Accurate Modeling of Molecular Atomization Energies with Machine Learning
• Longformer: The Long-Document Transformer
• DeePMD-kit: A deep learning package for many-body potential energy representation and molecular dynamics
• Interpretable machine learning: definitions, methods, and applications
• E(3)-Equivariant Graph Neural Networks for Data-Efficient and Accurate Interatomic Potentials
• Graph Networks as a Universal Machine Learning Framework for Molecules and Crystals
• Quantum-Chemical Insights from Deep Tensor Neural Networks
• Flamingo: a Visual Language Model for Few-Shot Learning
• Recent Advances and Applications of Deep Learning Methods in Materials Science
• Zero-Shot Text-to-Image Generation
• PhysNet: A Neural Network for Predicting Energies, Forces, Dipole Moments and Partial Charges
• SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systems
• DP-GEN: A concurrent learning platform for the generation of reliable deep learning based potential energy models
• BEiT: BERT Pre-Training of Image Transformers
• BLIP: Bootstrapping Language-Image Pre-training for Unified Vision-Language Understanding and Generation
• DScribe: Library of Descriptors for Machine Learning in Materials Science
• SISSO: a compressed-sensing method for identifying the best low-dimensional descriptor in an immensity of offered candidates
• Atomistic Line Graph Neural Network for Improved Materials Property Predictions
• Deep Graph Library: A Graph-Centric, Highly-Performant Package for Graph Neural Networks
• GNNExplainer: Generating Explanations for Graph Neural Networks
• Universal Fragment Descriptors for Predicting Electronic Properties of Inorganic Crystals
• Finding Density Functionals with Machine Learning
• FedKD: Communication Efficient Federated Learning via Knowledge Distillation
• A rigorous and robust quantum speed-up in supervised machine learning
• The Joint Automated Repository for Various Integrated Simulations (JARVIS) for data-driven materials design
• Do As I Can, Not As I Say: Grounding Language in Robotic Affordances
• CoCa: Contrastive Captioners are Image-Text Foundation Models
• Discriminative Embeddings of Latent Variable Models for Structured Data
• How to represent crystal structures for machine learning: towards fast prediction of electronic properties
• Opportunities and Challenges for Machine Learning in Materials Science
• Decision Transformer: Reinforcement Learning via Sequence Modeling
• GPUMD: A package for constructing accurate machine-learned potentials and performing highly efficient atomistic simulations
• Machine learning modeling of superconducting critical temperature
• Predicting materials properties without crystal structure: Deep representation learning from stoichiometry
• Interpretable and Explainable Machine Learning for Materials Science and Chemistry
• Benchmarking Materials Property Prediction Methods: The Matbench Test Set and Automatminer Reference Algorithm
• Directional Message Passing for Molecular Graphs
• Insightful classification of crystal structures using deep learning
• SpookyNet: Learning Force Fields with Electronic Degrees of Freedom and Nonlocal Effects
• Rethinking Network Design and Local Geometry in Point Cloud: A Simple Residual MLP Framework
• Combinatorial Optimization with Graph Convolutional Networks and Guided Tree Search
• WACSF - Weighted Atom-Centered Symmetry Functions as Descriptors in Machine Learning Potentials
• Few-Shot Parameter-Efficient Fine-Tuning is Better and Cheaper than In-Context Learning
• SELFIES and the future of molecular string representations
• Developing an improved Crystal Graph Convolutional Neural Network framework for accelerated materials discovery
• Machine Learning Enabled Computational Screening of Inorganic Solid Electrolytes for Dendrite Suppression with Li Metal Anode
• Chemical Shifts in Molecular Solids by Machine Learning
• Global Attention based Graph Convolutional Neural Networks for Improved Materials Property Prediction
• Equivariant message passing for the prediction of tensorial properties and molecular spectra
• Persistent homology analysis of protein structure, flexibility and folding
• Applying machine learning techniques to predict the properties of energetic materials
• OrbNet: Deep Learning for Quantum Chemistry Using Symmetry-Adapted Atomic-Orbital Features
• Recent advances on thermoelectric materials
• Perspective on integrating machine learning into computational chemistry and materials science
• Inclusion of machine learning kernel ridge regression potential energy surfaces in on-the-fly nonadiabatic molecular dynamics simulation
• Few-Shot Graph Learning for Molecular Property Prediction
• Graph Dynamical Networks for Unsupervised Learning of Atomic Scale Dynamics in Materials
• Learning from Few Examples: A Summary of Approaches to Few-Shot Learning
• Learning physical descriptors for materials science by compressed sensing
• Paired Open-Ended Trailblazer (POET): Endlessly Generating Increasingly Complex and Diverse Learning Environments and Their Solutions
• Discovering the Building Blocks of Atomic Systems using Machine Learning
• A sparse representation for potential energy surface
• Unsupervised machine learning via transfer learning and k-means clustering to classify materials image data
• Pretrained Transformers as Universal Computation Engines
• Artificial Intelligent Atomic Force Microscope Enabled by Machine Learning
• How to Fine-Tune BERT for Text Classification?
• Explainability in Graph Neural Networks: A Taxonomic Survey
• On Explainability of Graph Neural Networks via Subgraph Explorations
• Design and Analysis of Machine Learning Exchange-Correlation Functionals via Rotationally Invariant Convolutional Descriptors
• Accelerating amorphous polymer electrolyte screening by learning to reduce errors in molecular dynamics simulated properties
• Transformer Memory as a Differentiable Search Index
• Overcoming data scarcity with transfer learning
• AtomSets -- A Hierarchical Transfer Learning Framework for Small and Large Materials Datasets
• AutoPrompt: Eliciting Knowledge from Language Models with Automatically Generated Prompts
• Machine Learning and Computational Mathematics
• OpenPrompt: An Open-source Framework for Prompt-learning
• RODE: Learning Roles to Decompose Multi-Agent Tasks
• Fabrication of nanopores in a graphene sheet with heavy ions: a molecular dynamics study
• Lattice Dynamics and Thermal Equation of State of Platinum
• GraphVAMPNet, using graph neural networks and variational approach to markov processes for dynamical modeling of biomolecules
• Molecular Mechanics-Driven Graph Neural Network with Multiplex Graph for Molecular Structures
• Video PreTraining (VPT): Learning to Act by Watching Unlabeled Online Videos
• Harnessing Interpretable and Unsupervised Machine Learning to Address Big Data from Modern X-ray Diffraction
• Improving Attention Mechanism in Graph Neural Networks via Cardinality Preservation
• Dynamic GPU Energy Optimization for Machine Learning Training Workloads
• Building Machine Translation Systems for the Next Thousand Languages
• IRNet: A General Purpose Deep Residual Regression Framework for Materials Discovery
• Batched Multi-armed Bandits Problem
• Mastering Atari Games with Limited Data
• Extensible Structure-Informed Prediction of Formation Energy with Improved Accuracy and Usability employing Neural Networks
• Memory Replay with Data Compression for Continual Learning
• NITI: Training Integer Neural Networks Using Integer-only Arithmetic
• Molecule Identification with Rotational Spectroscopy and Probabilistic Deep Learning
• FuseDream: Training-Free Text-to-Image Generation with Improved CLIP+GAN Space Optimization
• Inversion of two-band superconductivity at the critical electron doping of (Mg,Al)B${}_2$
• Screening and understanding Li adsorption on 2-dimensional metallic materials by learning physics
• Topological descriptor of thermal conductivity in amorphous materials
• Strain engineering of ion migration in LiCoO2
• Don't Change the Algorithm, Change the Data: Exploratory Data for Offline Reinforcement Learning
• Possible complete miscibility of $(BN{)}_x(C_2{)}_{1-x}$ alloys
• Evolving Curricula with Regret-Based Environment Design
• Control of magnetic anisotropy by orbital hybridization in (La0.67Sr0.33MnO3)n/(SrTiO3)n superlattice
• Structural and electronic properties of Al nanowires: an ab initio pseudopotential study
• Beta-CROWN: Efficient Bound Propagation with Per-neuron Split Constraints for Complete and Incomplete Neural Network Robustness Verification
• Ab initio study of anisotropic mechanical and electronic properties of strained carbon-nitride nanosheet with interlayer bonding
• Tuning the Magnetic and Electronic Properties of Strontium Titanate by Carbon Doping
• Driver2vec: Driver Identification from Automotive Data
• Trends in bonding configuration at SiC/III-V semiconductor interfaces
• PointNorm: Dual Normalization is All You Need for Point Cloud Analysis
• Coarse-to-Fine Q-attention with Learned Path Ranking
• Sim-to-Real 6D Object Pose Estimation via Iterative Self-training for Robotic Bin Picking
• Preferential Temporal Difference Learning
• Challenges and opportunities for machine learning in multiscale computational modeling
• Imitating, Fast and Slow: Robust learning from demonstrations via decision-time planning

Cited by in corpus (4)

• Artificial Intelligence in Materials Science and Engineering: Current Landscape, Key Challenges, and Future Trajectorie
• Active and transfer learning with partially Bayesian neural networks for materials and chemicals
• Efficient dataset generation for machine learning perovskite alloys
• Enhancing Large Language Models with Domain-Specific Knowledge: The Case in Topological Materials