A General Tensor Prediction Framework Based on Graph Neural Networks

Article Multidisciplinary Sciences

Learning local equivariant representations for large-scale atomistic dynamics

Albert Musaelian et al.

Summary: This study introduces Allegro, a local equivariant deep neural network interatomic potential architecture that achieves excellent accuracy and scalability in quantum chemistry and molecular simulations.

NATURE COMMUNICATIONS (2023)

添加到收藏夹

Article Chemistry, Physical

Thermally Averaged Magnetic Anisotropy Tensors via Machine Learning Based on Gaussian Moments

Viktor Zaverkin et al.

Summary: The study introduces a machine learning method based on the Gaussian moment neural network for modeling molecular tensorial quantities, specifically the magnetic anisotropy tensor, achieving high accuracy and generalization capability. By combining machine-learned interatomic potential energies, the approach can be used to investigate the dynamic behavior of magnetic anisotropy tensors and offer unique insights into spin-phonon relaxation.

JOURNAL OF CHEMICAL THEORY AND COMPUTATION (2022)

添加到收藏夹

Article Chemistry, Multidisciplinary

Discovery of Lead-Free Perovskites for High-Performance Solar Cells via Machine Learning: Ultrabroadband Absorption, Low Radiative Combination, and Enhanced Thermal Conductivities

Xia Cai et al.

Summary: The authors propose a multi-step and multi-stage screening scheme to accelerate the discovery of hybrid organic-inorganic perovskites, combining machine learning with high-throughput calculations for pursuing excellent efficiency and thermal stability in solar cells. Through this screening process, they successfully identified four experimental-feasible candidates with good stability and high efficiency, which were further verified by density-functional theory calculations.

ADVANCED SCIENCE (2022)

添加到收藏夹

Review Chemistry, Multidisciplinary

Machine Learning: An Advanced Platform for Materials Development and State Prediction in Lithium-Ion Batteries

Chade Lv et al.

Summary: AI technology and computational chemistry can accelerate the research and development of novel battery systems, with successful examples, challenges of deploying AI in real-world scenarios, and an integrated framework outlined. The applications of machine learning in property prediction and battery discovery, as well as the prediction of battery states, are further summarized.

ADVANCED MATERIALS (2022)

添加到收藏夹

Article Multidisciplinary Sciences

E(3)-equivariant graph neural networks for data-efficient and accurate interatomic potentials

Simon Batzner et al.

Summary: This paper introduces an E(3)-equivariant deep learning method for accelerating molecular dynamics simulations. The method shows state-of-the-art accuracy and remarkable sample efficiency in faithfully describing the dynamics of complex systems. The Neural Equivariant Interatomic Potentials (NequIP) approach employs E(3)-equivariant convolutions to interact with geometric tensors, resulting in a more information-rich and faithful representation of atomic environments. NequIP outperforms existing models with significantly fewer training data, challenging the commonly held belief about the necessity of massive training sets for deep neural networks.

NATURE COMMUNICATIONS (2022)

添加到收藏夹

Article Materials Science, Multidisciplinary

Predicting tensorial molecular properties with equivariant machine learning models

Vu Ha Anh Nguyen et al.

Summary: Embedding molecular symmetries into machine learning models is key for efficient learning of chemico-physical scalar properties. In this work, the authors propose a scalable equivariant machine learning model based on local atomic environment descriptors for accurate predictions of dielectric and magnetic tensorial properties of different ranks in molecules.

PHYSICAL REVIEW B (2022)

添加到收藏夹

Review Chemistry, Multidisciplinary

Data-Driven Strategies for Accelerated Materials Design

Robert Pollice et al.

Summary: The ongoing revolution in the natural sciences, driven by machine learning and artificial intelligence, has sparked significant interest in the material science community. To address the imminent climate catastrophe and transform to clean energy, the development of new materials is crucial. New materials, such as organic photovoltaics, show great potential in replacing existing materials and opening up new applications.

ACCOUNTS OF CHEMICAL RESEARCH (2021)

添加到收藏夹

Article Chemistry, Physical

Challenges for machine learning force fields in reproducing potential energy surfaces of flexible molecules

Valentin Vassilev-Galindo et al.

Summary: The study evaluates the performance of machine learning models in reproducing complex potential-energy surfaces, finding that predictions greatly depend on the ML method used and the local region of the PES being sampled. It suggests switching from learning the entire PES within a single model to using multiple local models for different parts of the complex PES.

JOURNAL OF CHEMICAL PHYSICS (2021)

添加到收藏夹

Article Chemistry, Physical

Improving molecular force fields across configurational space by combining supervised and unsupervised machine learning

Gregory Fonseca et al.

Summary: In this study, a combination of unsupervised and supervised machine learning methods was used to bypass the bias of data for common configurations, which resulted in a significant decrease in prediction errors for force on non-equilibrium geometries. The approach also demonstrated improved stability over default training methods, allowing reliable study of processes involving highly out-of-equilibrium molecular configurations. These findings were consistent across different types of machine learning models used in the study.

JOURNAL OF CHEMICAL PHYSICS (2021)

添加到收藏夹

Review Chemistry, Physical

Machine Learning in Screening High Performance Electrocatalysts for CO2 Reduction

Ning Zhang et al.

Summary: This paper reviews the development of machine learning methods in screening CO2 reduction electrocatalysts, which can predict and understand the catalytic activity and reaction pathway of various materials. Machine learning is recognized as a fast and low-cost method for effectively exploring high performance electrocatalysts for CO2 reduction.

SMALL METHODS (2021)

添加到收藏夹

Article Chemistry, Physical

Cartesian message passing neural networks for directional properties: Fast and transferable atomic multipoles

Zachary L. Glick et al.

Summary: The study introduces a modified Cartesian MPNN framework for predicting atomic multipoles, demonstrating its accuracy and effectiveness on a large-scale dataset. The model accurately predicts atom-centered charges, dipoles, and quadrupoles, with minimal impact on multipole-multipole electrostatic energies, and is able to model the conformational dependencies of a molecule.

JOURNAL OF CHEMICAL PHYSICS (2021)

添加到收藏夹

Review Nanoscience & Nanotechnology

Machine learning for alloys

Gus L. W. Hart et al.

Summary: The integration of machine learning and alloys has played a crucial role in advancing research on various materials. Computational materials science has benefited from advancements in machine learning methods and data generation, opening up new possibilities for alloy research.

NATURE REVIEWS MATERIALS (2021)

添加到收藏夹

Review Chemistry, Physical

Machine Learning Boosting the Development of Advanced Lithium Batteries

Yangting Liu et al.

Summary: The application of machine learning in the field of lithium batteries can accelerate the discovery of materials and predict performances, promoting the development of advanced lithium batteries.

SMALL METHODS (2021)

添加到收藏夹

Review Chemistry, Multidisciplinary

Machine Learning Approaches for Thermoelectric Materials Research

Tian Wang et al.

ADVANCED FUNCTIONAL MATERIALS (2020)

添加到收藏夹

Article Materials Science, Multidisciplinary

Opportunities and Challenges for Machine Learning in Materials Science

Dane Morgan et al.

Annual Review of Materials Research (2020)

添加到收藏夹

Article Chemistry, Physical

High-throughput density functional perturbation theory and machine learning predictions of infrared, piezoelectric, and dielectric responses

Kamal Choudhary et al.

NPJ COMPUTATIONAL MATERIALS (2020)

添加到收藏夹

Article Materials Science, Multidisciplinary

Developing an improved crystal graph convolutional neural network framework for accelerated materials discovery

Cheol Woo Park et al.

PHYSICAL REVIEW MATERIALS (2020)

添加到收藏夹

Review Chemistry, Multidisciplinary

Learning to Make Chemical Predictions: The Interplay of Feature Representation, Data, and Machine Learning Methods

Mojtaba Haghighatlari et al.

CHEM (2020)

添加到收藏夹

Review Chemistry, Physical

Machine Learning for Atomic Simulation and Activity Prediction in Heterogeneous Catalysis: Current Status and Future

Sicong Ma et al.

ACS CATALYSIS (2020)

添加到收藏夹

Article Chemistry, Physical

PhysNet: A Neural Network for Predicting Energies, Forces, Dipole Moments, and Partial Charges

Oliver T. Unke et al.

JOURNAL OF CHEMICAL THEORY AND COMPUTATION (2019)

添加到收藏夹

Article Physics, Applied

Prediction of dielectric constants using a combination of first principles calculations and machine learning

Yuji Umeda et al.

JAPANESE JOURNAL OF APPLIED PHYSICS (2019)

添加到收藏夹

Article Computer Science, Interdisciplinary Applications

DeePMD-kit: A deep learning package for many-body potential energy representation and molecular dynamics

Han Wang et al.

COMPUTER PHYSICS COMMUNICATIONS (2018)

添加到收藏夹

Article Chemistry, Physical

SchNet - A deep learning architecture for molecules and materials

K. T. Schuett et al.

JOURNAL OF CHEMICAL PHYSICS (2018)

添加到收藏夹

Article Chemistry, Physical

Alchemical and structural distribution based representation for universal quantum machine learning

Felix A. Faber et al.

JOURNAL OF CHEMICAL PHYSICS (2018)

添加到收藏夹

Article Chemistry, Physical

A reactive, scalable, and transferable model for molecular energies from a neural network approach based on local information

Oliver T. Unke et al.

JOURNAL OF CHEMICAL PHYSICS (2018)

添加到收藏夹

Article Physics, Multidisciplinary

Symmetry-Adapted Machine Learning for Tensorial Properties of Atomistic Systems

Andrea Grisafi et al.

PHYSICAL REVIEW LETTERS (2018)

添加到收藏夹

Article Physics, Multidisciplinary

Crystal Graph Convolutional Neural Networks for an Accurate and Interpretable Prediction of Material Properties

Tian Xie et al.

PHYSICAL REVIEW LETTERS (2018)

添加到收藏夹

Review Multidisciplinary Sciences

Machine learning for molecular and materials science

Keith T. Butler et al.

NATURE (2018)

添加到收藏夹

Article Multidisciplinary Sciences

Accelerated discovery of stable lead-free hybrid organic-inorganic perovskites via machine learning

Shuaihua Lu et al.

NATURE COMMUNICATIONS (2018)

添加到收藏夹

Review Materials Science, Multidisciplinary

Review and perspective on ferroelectric HfO2-based thin films for memory applications

Min Hyuk Park et al.

MRS COMMUNICATIONS (2018)

添加到收藏夹

Article Multidisciplinary Sciences

Quantum-chemical insights from deep tensor neural networks

Kristof T. Schuett et al.

NATURE COMMUNICATIONS (2017)

添加到收藏夹

Article Engineering, Electrical & Electronic

A computational study of hafnia-based ferroelectric memories: from ab initio via physical modeling to circuit models of ferroelectric device

Milan Pesic et al.

JOURNAL OF COMPUTATIONAL ELECTRONICS (2017)

添加到收藏夹

Article Multidisciplinary Sciences

Machine learning of accurate energy-conserving molecular force fields

Stefan Chmiela et al.

SCIENCE ADVANCES (2017)

添加到收藏夹

Article Materials Science, Multidisciplinary

On representing chemical environments

Albert P. Bartok et al.

PHYSICAL REVIEW B (2013)

添加到收藏夹

Article Physics, Applied

Ferroelectricity in hafnium oxide thin films

T. S. Boescke et al.

APPLIED PHYSICS LETTERS (2011)

添加到收藏夹

Article Physics, Multidisciplinary

Generalized neural-network representation of high-dimensional potential-energy surfaces

Joerg Behler et al.

PHYSICAL REVIEW LETTERS (2007)

添加到收藏夹

Article Physics, Applied

Thermodynamic stability of high-K dielectric metal oxides ZrO2 and HfO2 in contact with Si and SiO2

M Gutowski et al.

APPLIED PHYSICS LETTERS (2002)

添加到收藏夹

A General Tensor Prediction Framework Based on Graph Neural Networks

相关参考文献

导出引文

分享论文