scholarly journals Optimizing many-body atomic descriptors for enhanced computational performance of machine learning based interatomic potentials

2019 ◽  
Vol 100 (2) ◽  
Author(s):  
Miguel A. Caro
Keyword(s):  
Machine Learning ◽  
Interatomic Potentials ◽  
Many Body ◽  
Computational Performance ◽  
Atomic Descriptors

scholarly journals A transferable active-learning strategy for reactive molecular force fields

2021 ◽  
Author(s):  
Tom Young ◽  
Tristan Johnston-Wood ◽  
Volker L. Deringer ◽  
Fernanda Duarte
Keyword(s):  
Machine Learning ◽  
Active Learning ◽  
Learning Strategy ◽  
Force Fields ◽  
Molecular Simulations ◽  
Quantum Mechanical ◽  
Interatomic Potentials ◽  
Molecular Force ◽  
High Level ◽  
Active Learning Strategy

Predictive molecular simulations require fast, accurate and reactive interatomic potentials. Machine learning offers a promising approach to construct such potentials by fitting energies and forces to high-level quantum-mechanical data, but...

scholarly journals Machine learning outperforms thermodynamics in measuring how well a many-body system learns a drive

2021 ◽  
Vol 11 (1) ◽  
Author(s):  
Weishun Zhong ◽  
Jacob M. Gold ◽  
Sarah Marzen ◽  
Jeremy L. England ◽  
Nicole Yunger Halpern
Keyword(s):  
Machine Learning ◽  
Novelty Detection ◽  
Representation Learning ◽  
Many Body ◽  
Discrimination Ability ◽  
Machine Learning Model ◽  
Systems Learning ◽  
Statistical Mechanical ◽  
Far From Equilibrium ◽  
Many Body Systems

AbstractDiverse many-body systems, from soap bubbles to suspensions to polymers, learn and remember patterns in the drives that push them far from equilibrium. This learning may be leveraged for computation, memory, and engineering. Until now, many-body learning has been detected with thermodynamic properties, such as work absorption and strain. We progress beyond these macroscopic properties first defined for equilibrium contexts: We quantify statistical mechanical learning using representation learning, a machine-learning model in which information squeezes through a bottleneck. By calculating properties of the bottleneck, we measure four facets of many-body systems’ learning: classification ability, memory capacity, discrimination ability, and novelty detection. Numerical simulations of a classical spin glass illustrate our technique. This toolkit exposes self-organization that eludes detection by thermodynamic measures: Our toolkit more reliably and more precisely detects and quantifies learning by matter while providing a unifying framework for many-body learning.

scholarly journals Taking materials dynamics to new extremes using machine learning interatomic potentials

2021 ◽  
Author(s):  
Yang Yang ◽  
Long Zhao ◽  
Chen-Xu Han ◽  
Xiang-Dong Ding ◽  
Turab Lookman ◽  
...  
Keyword(s):  
Machine Learning ◽  
Interatomic Potentials

Dual adaptive sampling and machine learning interatomic potentials for modeling materials with chemical bond hierarchy

2021 ◽  
Vol 104 (9) ◽  
Author(s):  
Hongliang Yang ◽  
Yifan Zhu ◽  
Erting Dong ◽  
Yabei Wu ◽  
Jiong Yang ◽  
...  
Keyword(s):  
Machine Learning ◽  
Chemical Bond ◽  
Adaptive Sampling ◽  
Interatomic Potentials

Machine Learning Interatomic Potentials for Global Optimization and Molecular Dynamics Simulation

2019 ◽  
pp. 253-288 ◽  
Author(s):  
Ivan A. Kruglov ◽  
Pavel E. Dolgirev ◽  
Artem R. Oganov ◽  
Arslan B. Mazitov ◽  
Sergey N. Pozdnyakov ◽  
...  
Keyword(s):  
Machine Learning ◽  
Molecular Dynamics ◽  
Global Optimization ◽  
Molecular Dynamics Simulation ◽  
Dynamics Simulation ◽  
Interatomic Potentials

Accessing thermal conductivity of complex compounds by machine learning interatomic potentials

2019 ◽  
Vol 100 (14) ◽  
Author(s):  
Pavel Korotaev ◽  
Ivan Novoselov ◽  
Aleksey Yanilkin ◽  
Alexander Shapeev
Keyword(s):  
Machine Learning ◽  
Thermal Conductivity ◽  
Complex Compounds ◽  
Interatomic Potentials

Machine learning prediction of interaction energies in rigid water clusters

2018 ◽  
Vol 20 (35) ◽  
pp. 22987-22996 ◽  
Author(s):  
Samik Bose ◽  
Diksha Dhawan ◽  
Sutanu Nandi ◽  
Ram Rup Sarkar ◽  
Debashree Ghosh
Keyword(s):  
Machine Learning ◽  
Support Vector Regression ◽  
Water Clusters ◽  
High Accuracy ◽  
Support Vector ◽  
Interaction Energies ◽  
Many Body ◽  
New Machine

A new machine learning based approach combining support vector regression (SVR) and many body expansion (MBE) that can predict the interaction energies of water clusters with high accuracy (for decamers: 2.78% of QM estimates).

scholarly journals Performance and Cost Assessment of Machine Learning Interatomic Potentials

2020 ◽  
Vol 124 (4) ◽  
pp. 731-745 ◽  
Author(s):  
Yunxing Zuo ◽  
Chi Chen ◽  
Xiangguo Li ◽  
Zhi Deng ◽  
Yiming Chen ◽  
...  
Keyword(s):  
Machine Learning ◽  
Cost Assessment ◽  
Interatomic Potentials

scholarly journals Machine Learning Classical Interatomic Potentials for Molecular Dynamics from First-Principles Training Data

2019 ◽  
Vol 123 (12) ◽  
pp. 6941-6957 ◽  
Author(s):  
Henry Chan ◽  
Badri Narayanan ◽  
Mathew J. Cherukara ◽  
Fatih G. Sen ◽  
Kiran Sasikumar ◽  
...  
Keyword(s):  
Machine Learning ◽  
Molecular Dynamics ◽  
First Principles ◽  
Training Data ◽  
Interatomic Potentials

Molecular Dynamics Simulations of Atoms Diffusion in Solid

2018 ◽  
Vol 15 ◽  
pp. 51-64
Author(s):  
Yu Lu Zhou ◽  
Xiao Ma Tao ◽  
Qing Hou ◽  
Yi Fang Ouyang
Keyword(s):  
Molecular Dynamics ◽  
Diffusion Coefficients ◽  
Md Simulations ◽  
Einstein Relation ◽  
Interatomic Potentials ◽  
Body System ◽  
Many Body ◽  
Point Particles ◽  
Bulk Surface ◽  
Dynamics Simulations

Molecular dynamics (MD) simulations, which treat atoms as point particles and trace their individual trajectories, are always employed to investigate the transport properties of a many-body system. The diffusion coefficients of atoms in solid can be obtained by the Einstein relation and the Green-Kubo relation. An overview of the MD simulations of atoms diffusion in the bulk, surface and grain boundary is provided. We also give an example of the diffusion of helium in tungsten to illustrate the procedure, as well as the importance of the choice of interatomic potentials. MD simulations can provide intuitive insights into the atomic mechanisms of diffusion.

Sign in / Sign up

Export Citation Format

Share Document