Recent advances and applications of deep learning methods in materials science (2021-10-28T00:00:00.000000Z)

TL;DR

A high-level overview of deep learning methods followed by a detailed discussion of recent developments ofdeep learning in atomistic simulation, materials imaging, spectral analysis, and natural language processing is presented.

Abstract

Deep learning (DL) is one of the fastest-growing topics in materials data science, with rapidly emerging applications spanning atomistic, image-based, spectral, and textual data modalities. DL allows analysis of unstructured data and automated identification of features. The recent development of large materials databases has fueled the application of DL methods in atomistic prediction in particular. In contrast, advances in image and spectral data have largely leveraged synthetic data enabled by high-quality forward models as well as by generative unsupervised DL methods. In this article, we present a high-level overview of deep learning methods followed by a detailed discussion of recent developments of deep learning in atomistic simulation, materials imaging, spectral analysis, and natural language processing. For each modality we discuss applications involving both theoretical and experimental data, typical modeling approaches with their strengths and limitations, and relevant publicly available software and datasets. We conclude the review with a discussion of recent cross-cutting work related to uncertainty quantification in this field and a brief perspective on limitations, challenges, and potential growth areas for DL methods in materials science.

Authors

Anubhav Jain

4 papers

Chi Chen

3 papers

S. Ong

5 papers

TL;DR

Abstract

Authors

References390 items

Prediction of the Electron Density of States for Crystalline Compounds with Atomistic Line Graph Neural Networks (ALIGNN)

Graph Neural Network Predictions of Metal Organic Framework CO2 Adsorption Properties

Cross-property deep transfer learning framework for enhanced predictive analytics on small materials data

Perceptrons

Density of states prediction for materials discovery via contrastive learning from probabilistic embeddings

Graph convolutional network for predicting abnormal grain growth in Monte Carlo simulations of microstructural evolution

Artificial intelligence for search and discovery of quantum materials

ReaxFF-MPNN machine learning potential: a combination of reactive force field and message passing neural networks.

Generative models for inverse design of inorganic solid materials

Deep Learning-Based Prediction of Material Properties Using Chemical Compositions and Diffraction Patterns as Experimentally Accessible Inputs.

Automating crystal-structure phase mapping by combining deep learning with constraint reasoning

Addressing materials’ microstructure diversity using transfer learning

Autonomous experimentation systems for materials development: A community perspective

Transformer-based artificial neural networks for the conversion between chemical notations

DGL-LifeSci: An Open-Source Toolkit for Deep Learning on Graphs in Life Science

Database of ab initio L-edge X-ray absorption near edge structure

Atomistic Line Graph Neural Network for improved materials property predictions

Deep Learning‐Assisted Quantification of Atomic Dopants and Defects in 2D Materials

Best practices in machine learning for chemistry

Materials Representation and Transfer Learning for Multi-Property Prediction

A deep convolutional neural network for real-time full profile analysis of big powder diffraction data

Accurate and scalable graph neural network force field and molecular dynamics with direct force architecture

The abTEM code: transmission electron microscopy from first principles

Toward autonomous design and synthesis of novel inorganic materials.

Learning Multiple Quantiles With Neural Networks

Validation of non-negative matrix factorization for rapid assessment of large sets of atomic pair distribution function data

Accelerating materials discovery with Bayesian optimization and graph deep learning

Inverse design of crystal structures for multicomponent systems

Ensemble deep learning: A review

OPTIMADE, an API for exchanging materials data

Phase-Transition-Enhanced Thermoelectric Transport in Rickardite Mineral Cu3–xTe2

Enabling deeper learning on big data for materials informatics applications

Computational scanning tunneling microscope image database

Opportunities and challenges of text mining in aterials research

High-throughput search for magnetic topological materials using spin-orbit spillage, machine learning, and experiments

A General Framework Combining Generative Adversarial Networks and Mixture Density Networks for Inverse Modeling in Microstructural Materials Design

Benchmarking graph neural networks for materials chemistry

GNINA 1.0: molecular docking with deep learning

DeepFrag: a deep convolutional neural network for fragment-based lead optimization

Plasma Hsp90 levels in patients with systemic sclerosis and relation to lung and skin involvement: a cross-sectional and longitudinal study

Instance Segmentation for Direct Measurements of Satellites in Metal Powders and Automated Microstructural Characterization from Image Data

Machine learned features from density of states for accurate adsorption energy prediction

Theoretical prediction of high melting temperature for a Mo–Ru–Ta–W HCP multiprincipal element alloy

Anomalous collapses of Nares Strait ice arches leads to enhanced export of Arctic sea ice

A cloud platform for atomic pair distribution function analysis: PDFitc.

Learning properties of ordered and disordered materials from multi-fidelity data

TorchMD: A Deep Learning Framework for Molecular Simulations

Leveraging Uncertainty from Deep Learning for Trustworthy Material Discovery Workflows

Crystal symmetry classification from powder X-ray diffraction patterns using a convolutional neural network

Layer-wise anomaly detection and classification for powder bed additive manufacturing processes: A machine-agnostic algorithm for real-time pixel-wise semantic segmentation

Symmetry prediction and knowledge discovery from X-ray diffraction patterns using an interpretable machine learning approach

Data-driven materials research enabled by natural language processing and information extraction

Diversity oriented Deep Reinforcement Learning for targeted molecule generation

Machine learning in materials genome initiative: A review

A Review of Uncertainty Quantification in Deep Learning: Techniques, Applications and Challenges

DeepFrag-k: a fragment-based deep learning approach for protein fold recognition

Machine learning the quantum-chemical properties of metal–organic frameworks for accelerated materials discovery

Active Reaction Control of Cu Redox State Based on Real-Time Feedback from In Situ Synchrotron Measurements.

The Open Catalyst 2020 (OC20) Dataset and Community Challenges

Effect of electron-phonon scattering, pressure and alloying on the thermoelectric performance of TmCu$_3$Ch$_4$ (Tm=V, Nb, Ta; Ch=S, Se, Te)

Compact representations of microstructure images using triplet networks

An Introduction to Electrocatalyst Design using Machine Learning for Renewable Energy Storage

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

Using word embeddings in abstracts to accelerate metallocene catalysis polymerization research

A fourth-generation high-dimensional neural network potential with accurate electrostatics including non-local charge transfer

Direct Prediction of Phonon Density of States With Euclidean Neural Networks

A database of battery materials auto-generated using ChemDataExtractor

Crystallography companion agent for high-throughput materials discovery

Deep learning analysis on microscopic imaging in materials science

Constrained crystals deep convolutional generative adversarial network for the inverse design of crystal structures

Accelerated mapping of electronic density of states patterns of metallic nanoparticles via machine-learning

Reproducibility in high-throughput density functional theory: a comparison of AFLOW, Materials Project, and OQMD

The joint automated repository for various integrated simulations (JARVIS) for data-driven materials design

Qualitative Analysis of Monte Carlo Dropout

Opportunities and Challenges for Machine Learning in Materials Science

Microstructure Generation via Generative Adversarial Network for Heterogeneous, Topologically Complex 3D Materials