☆ 4.7 Article

Robustness of Local Predictions in Atomistic Machine Learning Models

JOURNAL OF CHEMICAL THEORY AND COMPUTATION (2023)

相关参考文献

注意：仅列出部分参考文献，下载原文获取全部文献信息。

Article Chemistry, Physical

A short blanket dilemma for a state-of-the-art neural network potential for water: Reproducing experimental properties or the physics of the underlying many-body interactions?

Yaoguang Zhai et al.

Summary: Deep neural network (DNN) potentials have gained popularity in computer simulations of molecular systems. This study combines the computational efficiency of the DeePMD framework and the accuracy of the MB-pol potential to train a DNN potential for large-scale simulations of water. The DNN potential can reproduce MB-pol results for liquid water but lacks accuracy for vapor-liquid equilibrium properties. Attempts to include many-body effects result in a new DNN potential that accurately reproduces vapor-liquid equilibrium properties but loses accuracy in liquid properties.

JOURNAL OF CHEMICAL PHYSICS (2023)

添加到收藏夹

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, Multidisciplinary

Frustration in Super-Ionic Conductors Unraveled by the Density of Atomistic States

Shuo Wang et al.

Summary: The frustration in super-ionic conductors enables their high ionic conductivities, which are important for various technological applications. However, analyzing the disordering of atomistic configurations is challenging. By using lithium super-ionic conductors as models, the density of atomistic states (DOAS) analytics is proposed and demonstrated to quantitatively characterize disorder, reveal the energetics of local disorder, and understand how frustration enhances diffusion. Material design strategies aided by DOAS are also devised and demonstrated for new super-ionic conductors. DOAS is a versatile analytic tool for understanding complex atomistic systems and guiding material design.

ANGEWANDTE CHEMIE-INTERNATIONAL EDITION (2023)

添加到收藏夹

Article Materials Science, Multidisciplinary

Modeling high-entropy transition metal alloys with alchemical compression

Nataliya Lopanitsyna et al.

Summary: Alloys composed of various elements in roughly equal proportion, known as high-entropy alloys, have attracted much attention due to their thermodynamics and unique mechanical properties, as well as their potential for catalytic applications. However, traditional atomistic modeling and data-driven potentials face challenges in dealing with the increasing number of elements. In this study, a new method is proposed to compress chemical information and reduce the computational cost, allowing the description of 25 transition metals. The accuracy and stability of the model are demonstrated by studying prototype alloys and simulating element segregation. The findings also provide insights for alloy design through the establishment of data-driven Hume-Rothery rules.

PHYSICAL REVIEW MATERIALS (2023)

添加到收藏夹

Article Chemistry, Physical

Temperature- and vacancy-concentration-dependence of heat transport in Li3ClO from multi-method numerical simulations

Paolo Pegolo et al.

Summary: This study calculates the thermal conductivity of the Li3ClO antiperovskite, a prototypical Li-ion conductor, using advanced methods and reveals the significant role of anharmonic interactions and diffusive defects.

NPJ COMPUTATIONAL MATERIALS (2022)

添加到收藏夹

Article Multidisciplinary Sciences

Efficient Calculation of the Lattice Thermal Conductivity by Atomistic Simulations with Ab Initio Accuracy

Joakim Brorsson et al.

Summary: High-order force constant expansions, combined with GPU-accelerated molecular dynamics simulations, provide an accurate, transferable, and efficient approach for sampling the dynamical properties of materials.

ADVANCED THEORY AND SIMULATIONS (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 Chemistry, Multidisciplinary

Automated Bonding Analysis with Crystal Orbital Hamilton Populations

Janine George et al.

Summary: Understanding crystalline structures based on chemical bonding is becoming increasingly important. This study presents a new set of tools for automating the calculation of Crystal Orbital Hamilton Population (COHP) and analyzing the results. The tools enable high-throughput bonding analysis and facilitate the use of bonding information for machine learning studies.

CHEMPLUSCHEM (2022)

添加到收藏夹

Article Materials Science, Multidisciplinary

Predicting hot-electron free energies from ground-state data

Chiheb Ben Mahmoud et al.

Summary: The article discusses the prediction of electronic free energy at arbitrary electron temperature using machine learning, avoiding the need to train temperature-dependent potentials, providing a new method. By combining physical considerations with machine-learning predictions, it offers a blueprint for the development of atomistic modeling.

PHYSICAL REVIEW B (2022)

添加到收藏夹

Article Multidisciplinary Sciences

Origins of structural and electronic transitions in disordered silicon

Volker L. Deringer et al.

Summary: This study investigates the structural transitions of amorphous silicon under increasing external pressure, revealing a three-step transformation sequence and demonstrating the transient nature of the VHDA phase, which rapidly nucleates crystallites. The machine learning model for electronic density of states provides support for the onset of metallicity during VHDA formation and subsequent crystallization, showcasing a machine learning-driven approach to predictive materials modeling.

NATURE (2021)

添加到收藏夹

Article Multidisciplinary Sciences

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

Tsz Wai Ko et al.

Summary: Machine learning potentials have limitations in accounting for long-range charge transfer, which can be overcome with the introduction of a fourth-generation high-dimensional neural network potential that includes non-local charge population information. This method provides forces, charges, and energies in excellent agreement with DFT data, significantly extending the applicability of modern machine learning potentials.

NATURE COMMUNICATIONS (2021)

添加到收藏夹

Article Chemistry, Physical

Thermal transport and phase transitions of zirconia by on-the-fly machine-learned interatomic potentials

Carla Verdi et al.

Summary: Machine-learned interatomic potentials are used to accurately calculate complex materials properties at finite temperatures with first-principles accuracy, but their ability to describe anharmonic properties is still uncertain. A new on-the-fly learning technique based on molecular dynamics and Bayesian inference is employed to generate an interatomic potential capable of describing the thermodynamic properties of zirconia. The study demonstrates that machine-learned potentials offer an accurate and efficient solution for simulating the thermodynamic properties of a wide range of anharmonic materials.

NPJ COMPUTATIONAL MATERIALS (2021)

添加到收藏夹

Article Materials Science, Multidisciplinary

Finite-temperature materials modeling from the quantum nuclei to the hot electron regime

Nataliya Lopanitsyna et al.

Summary: Atomistic simulations offer insights into structure-property relations at atomic scales while machine-learning techniques provide efficient approximations of accurate electronic structure calculations for computing finite-temperature properties. This combination allows for predicting material behavior at different temperatures with cost-effective accuracy in a wide range of applications.

PHYSICAL REVIEW MATERIALS (2021)

添加到收藏夹

Review Chemistry, Multidisciplinary

Gaussian Process Regression for Materials and Molecules

Volker L. Deringer et al.

Summary: This paper introduces the application of Gaussian process regression machine learning in computational materials science and chemistry, focusing on the regression of atomistic properties, including the construction of interatomic potentials, force fields, and fitting of various quantities. Methodological aspects of reference data generation, representation, regression, and model validation are reviewed, along with a survey of applications in chemistry and materials science and an outline of future development vision.

CHEMICAL REVIEWS (2021)

添加到收藏夹

Article Materials Science, Multidisciplinary

Accelerating finite-temperature Kohn-Sham density functional theory with deep neural networks

J. A. Ellis et al.

Summary: This study presents a numerical modeling workflow based on machine learning that can reproduce the total energies from Kohn-Sham density functional theory (DFT) at finite electronic temperature with high accuracy and low computational cost. Utilizing deep neural networks, the workflow provides the local density of states (LDOS) for a specific atomic configuration, enabling calculations of spatially resolved, energy-resolved, and integrated quantities, including the DFT total free energy. The efficacy of this approach is demonstrated for both solid and liquid metals, showing potential for multiscale materials modeling under different conditions.

PHYSICAL REVIEW B (2021)

添加到收藏夹

Article Chemistry, Multidisciplinary

Efficient and tunable one-dimensional charge transport in layered lanthanide metal-organic frameworks

Grigorii Skorupskii et al.

NATURE CHEMISTRY (2020)

添加到收藏夹

Article Chemistry, Multidisciplinary

LOBSTER: Local orbital projections, atomic charges, and chemical-bonding analysis fromprojector-augmented-wave-baseddensity-functional theory

Ryky Nelson et al.

JOURNAL OF COMPUTATIONAL CHEMISTRY (2020)

添加到收藏夹

Article Multidisciplinary Sciences

Materials Cloud, a platform for open computational science

Leopold Talirz et al.

SCIENTIFIC DATA (2020)

添加到收藏夹

Article Materials Science, Multidisciplinary

Learning the electronic density of states in condensed matter

Chiheb Ben Mahmoud et al.

PHYSICAL REVIEW B (2020)

添加到收藏夹

Article Chemistry, Inorganic & Nuclear

Rational modifications of PCN-700 to induce electrical conductivity: a computational study

Sanggyu Chong et al.

DALTON TRANSACTIONS (2020)

添加到收藏夹

Article Chemistry, Multidisciplinary

Quantifying Chemical Structure and Machine-Learned Atomic Energies in Amorphous and Liquid Silicon

Noam Bernstein et al.

ANGEWANDTE CHEMIE-INTERNATIONAL EDITION (2019)

添加到收藏夹

Article Chemistry, Physical

From Molecular Fragments to the Bulk: Development of a Neural Network Potential for MOF-5

Marco Eckhoff et al.

JOURNAL OF CHEMICAL THEORY AND COMPUTATION (2019)

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 Physics, Multidisciplinary

Growth Mechanism and Origin of High sp3 Content in Tetrahedral Amorphous Carbon

Miguel A. Caro et al.

PHYSICAL REVIEW LETTERS (2018)

添加到收藏夹

Article Physics, Multidisciplinary

Data-Driven Learning of Total and Local Energies in Elemental Boron

Volker L. Deringer et al.

PHYSICAL REVIEW LETTERS (2018)

添加到收藏夹

Article Chemistry, Physical

Computational Surface Chemistry of Tetrahedral Amorphous Carbon by Combining Machine Learning and Density Functional Theory

Volker L. Deringer et al.

CHEMISTRY OF MATERIALS (2018)

添加到收藏夹

Article Chemistry, Multidisciplinary

ANI-1: an extensible neural network potential with DFT accuracy at force field computational cost

J. S. Smith et al.

CHEMICAL SCIENCE (2017)

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

Review Chemistry, Multidisciplinary

Theoretical Treatment of CH3NH3PbI3 Perovskite Solar Cells

Sining Yun et al.

ANGEWANDTE CHEMIE-INTERNATIONAL EDITION (2017)

添加到收藏夹

Article Chemistry, Physical

Self-consistent implementation of meta-GGA functionals for the ONETEP linear-scaling electronic structure package

James C. Womack et al.

JOURNAL OF CHEMICAL PHYSICS (2016)

添加到收藏夹

Article Chemistry, Multidisciplinary

LOBSTER: A Tool to Extract Chemical Bonding from Plane-Wave Based DFT

Stefan Maintz et al.

JOURNAL OF COMPUTATIONAL CHEMISTRY (2016)

添加到收藏夹

Article Chemistry, Physical

Large-Scale Density Functional Theory Transition State Searching in Enzymes

Greg Lever et al.

JOURNAL OF PHYSICAL CHEMISTRY LETTERS (2014)

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 Materials Science, Multidisciplinary

Thermal transport in phase-change materials from atomistic simulations

Gabriele C. Sosso et al.

PHYSICAL REVIEW B (2012)

添加到收藏夹

Article Materials Science, Multidisciplinary

High-dimensional neural network potentials for metal surfaces: A prototype study for copper

Nongnuch Artrith et al.

PHYSICAL REVIEW B (2012)

添加到收藏夹

Article Materials Science, Multidisciplinary

Neural network interatomic potential for the phase change material GeTe

Gabriele C. Sosso et al.

PHYSICAL REVIEW B (2012)

添加到收藏夹

Review Chemistry, Multidisciplinary

Symmetry-adapted perturbation theory of intermolecular forces

Krzysztof Szalewicz

WILEY INTERDISCIPLINARY REVIEWS-COMPUTATIONAL MOLECULAR SCIENCE (2012)

添加到收藏夹

Article Chemistry, Physical

Electrostatic embedding in large-scale first principles quantum mechanical calculations on biomolecules

Stephen J. Fox et al.

JOURNAL OF CHEMICAL PHYSICS (2011)

添加到收藏夹

Article Materials Science, Multidisciplinary

Visualization and analysis of atomistic simulation data with OVITO-the Open Visualization Tool

Alexander Stukowski

MODELLING AND SIMULATION IN MATERIALS SCIENCE AND ENGINEERING (2010)

添加到收藏夹

Article Physics, Multidisciplinary

Gaussian Approximation Potentials: The Accuracy of Quantum Mechanics, without the Electrons

Albert P. Bartok et al.

PHYSICAL REVIEW LETTERS (2010)

添加到收藏夹

Article Materials Science, Multidisciplinary

Plane-wave based electronic structure calculations for correlated materials using dynamical mean-field theory and projected local orbitals

B. Amadon et al.

PHYSICAL REVIEW B (2008)

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 Chemistry, Physical

Linear scaling density fitting

Alex Sodt et al.

JOURNAL OF CHEMICAL PHYSICS (2006)

添加到收藏夹

Article Multidisciplinary Sciences

Nearsightedness of electronic matter

E Prodan et al.

PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA (2005)

添加到收藏夹

Article Chemistry, Physical

Improved Fermi operator expansion methods for fast electronic structure calculations

WZ Liang et al.

JOURNAL OF CHEMICAL PHYSICS (2003)

添加到收藏夹

Article Chemistry, Multidisciplinary

Sparse matrix multiplications for linear scaling electronic structure calculations in an atom-centered basis set using multiatom blocks

C Saravanan et al.

JOURNAL OF COMPUTATIONAL CHEMISTRY (2003)

添加到收藏夹

© Peeref 2019-2024. All rights reserved.