Filters








32 Hits in 8.6 sec

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

Han Wang, Linfeng Zhang, Jiequn Han, Weinan E
2018 Computer Physics Communications  
Here we describe DeePMD-kit, a package written in Python/C++ that has been designed to minimize the effort required to build deep learning based representation of potential energy and force field and to  ...  Recent developments in many-body potential energy representation via deep learning have brought new hopes to addressing the accuracy-versus-efficiency dilemma in molecular simulations.  ...  (GDML) [9] , and the Deep potential for molecular dynamics (DeePMD) [10, 11] .  ... 
doi:10.1016/j.cpc.2018.03.016 fatcat:vwbzjqwvwzdyfeshihk7aww4tm

Deep Potentials for Materials Science [article]

Tongqi Wen, Linfeng Zhang, Han Wang, Weinan E, David J. Srolovitz
2022 arXiv   pre-print
has emerged and been widely applied; i.e., machine learning potentials (MLPs).  ...  We also review materials applications of DPs in a wide range of materials systems. The DP Library provides a platform for the development of DPs and a database of extant DPs.  ...  Deep Potential Generator Deep Potential GENerator (DP-GEN) is a software package that implements the concurrent learning framework for generating high quality DPs. a. Basic concepts and framework.  ... 
arXiv:2203.00393v1 fatcat:s54mv67my5cyvdjfiir3w53bye

Pushing the limit of molecular dynamics with ab initio accuracy to 100 million atoms with machine learning [article]

Weile Jia, Han Wang, Mohan Chen, Denghui Lu, Lin Lin, Roberto Car, Weinan E, Linfeng Zhang
2020 arXiv   pre-print
We report that a machine learning-based simulation protocol (Deep Potential Molecular Dynamics), while retaining ab initio accuracy, can simulate more than 1 nanosecond-long trajectory of over 100 million  ...  atoms per day, using a highly optimized code (GPU DeePMD-kit) on the Summit supercomputer.  ...  Index Terms-Deep potential molecular dynamics, ab initio molecular dynamics, machine learning, GPU, heterogeneous architecture, Summit I.  ... 
arXiv:2005.00223v3 fatcat:yyl4enw2mrd5jn2uzamz4zzcmq

Choosing the right molecular machine learning potential

Max Pinheiro Jr, Fuchun Ge, Nicolas Ferré, Pavlo O. Dral, Mario Barbatti
2021 Chemical Science  
Quantum-chemistry simulations based on potential energy surfaces of molecules provide invaluable insight into the physicochemical processes at the atomistic level and yield such important observables as  ...  reaction rates and spectra....  ...  Stefan Grimme for helpful suggestions.  ... 
doi:10.1039/d1sc03564a pmid:34880991 pmcid:PMC8580106 fatcat:pcxp7zerlngqtep5bojb7h236y

Development of Interatomic Potential for Al-Tb Alloy by Deep Neural Network Learning Method [article]

L. Tang, Z. J. Yang, T. Q. Wen, K. M. Ho, M. J. Kramer, C. Z. Wang
2020 arXiv   pre-print
The atomic configurations and the corresponding total potential energies and forces on each atom obtained from ab initio molecular dynamics (AIMD) simulations are collected to train a DNN model to construct  ...  An interatomic potential for Al-Tb alloy around the composition of Al90Tb10 was developed using the deep neural network (DNN) learning method.  ...  Department of Energy (DOE), Office of Science, Basic Energy Sciences, Materials Science and Engineering Division  ... 
arXiv:2001.06762v3 fatcat:47yhexgjgjccvfovyjiunhbsmq

End-to-end Symmetry Preserving Inter-atomic Potential Energy Model for Finite and Extended Systems [article]

Linfeng Zhang, Jiequn Han, Han Wang, Wissam A. Saidi, Roberto Car, and Weinan E
2018 arXiv   pre-print
Machine learning models are changing the paradigm of molecular modeling, which is a fundamental tool for material science, chemistry, and computational biology.  ...  Here we develop Deep Potential - Smooth Edition (DeepPot-SE), an end-to-end machine learning-based PES model, which is able to efficiently represent the PES for a wide variety of systems with the accuracy  ...  Acknowledgments We thank the anonymous reviewers for their careful reading of our manuscript and insightful comments and suggestions. The work of L. Z., J. H., and W.  ... 
arXiv:1805.09003v2 fatcat:dwu6ppwvcvem7cernu5ttgq5ey

Atomic-scale measurement of localized dislocation phonons at Si/Ge interface [article]

Yuehui Li, Bo Han, Ruochen Shi, Ruishi Qi, Xiaorui Hao, Ning Li, Bingyao Liu, Jinlong Du, Ji Chen, Peng Gao
2022 arXiv   pre-print
Here, at the atomic scale we directly measure the localized phonon modes of dislocations at a Si/Ge interface using scanning transmission electron microscopy electron energy loss spectroscopy.  ...  Understanding such effects requires the knowledge of defect phonon modes, which however is largely unexplored in experiments due to the challenge in characterization of phonons for the atomic-sized defects  ...  We acknowledge the High-performance Computing Platform of Peking University for providing computational resources for the MD simulations.  ... 
arXiv:2206.02329v1 fatcat:q3rwfu7odvajvg6nt2bq4ov6tm

Deep Learning of Accurate Force Field of Ferroelectric HfO_2 [article]

Jing Wu, Yuzhi Zhang, Linfeng Zhang, Shi Liu
2020 arXiv   pre-print
Here we present a deep neural network-based interatomic force field of HfO_2 learned from ab initio data using a concurrent learning procedure.  ...  Molecular dynamics is an ideal technique for investigating dynamical processes on large length and time scales, though its applications to new materials is often hindered by the limited availability and  ...  In this work, the smooth version of the DP model was employed 72 and the DeePMD-kit package 76 was used for training.  ... 
arXiv:2010.16082v1 fatcat:col46mgz3va27mhkojflbj5l4u

Deep neural network for the dielectric response of insulators [article]

Linfeng Zhang, Mohan Chen, Xifan Wu, Han Wang, Weinan E, Roberto Car
2020 arXiv   pre-print
When combined with the Deep Potential model of the atomic potential energy surface, the scheme predicts the dielectric response of insulators for trajectories inaccessible to direct ab-initio simulation  ...  We demonstrate the approach by calculating the infrared spectra of liquid water at standard conditions, and of ice under extreme pressure, when it transforms from a molecular to an ionic crystal.  ...  The code for this work has been integrated into the open-source software package DeePMD-kit [51] and we used the DP-GEN package [52] for the iterative scheme.  ... 
arXiv:1906.11434v5 fatcat:yobdbijgmzeezpuq75btinozri

Machine Learning Force Fields

Oliver T. Unke, Stefan Chmiela, Huziel E. Sauceda, Michael Gastegger, Igor Poltavsky, Kristof T. Schütt, Alexandre Tkatchenko, Klaus-Robert Müller
2021 Chemical Reviews  
The key idea is to learn the statistical relation between chemical structure and potential energy without relying on a preconceived notion of fixed chemical bonds or knowledge about the relevant interactions  ...  The core concepts underlying ML-FFs are described in detail, and a step-by-step guide for constructing and testing them from scratch is given.  ...  We would like to thank Stefan Ganscha for the valuable input to the manuscript.  ... 
doi:10.1021/acs.chemrev.0c01111 pmid:33705118 pmcid:PMC8391964 fatcat:7zssmfreg5alnnrjcwvh4hipk4

Accurate force field of two-dimensional ferroelectrics from deep learning [article]

Jing Wu, Liyi Bai, Jiawei Huang, Liyang Ma, Jian Liu, Shi Liu
2021 arXiv   pre-print
Molecular dynamics (MD) using classical force fields is a reliable and efficient method for large-scale simulations of dynamical processes with atomic resolution.  ...  Here we developed a deep neural network-based force field of monolayer In_2Se_3 using a concurrent learning procedure that efficiently updates the first-principles-based training database.  ...  For a many-body potential such as REBO that uses a highly sophisticated energy function with a large number of parameters, the fitting process is often tedious.  ... 
arXiv:2109.07104v2 fatcat:fpeh6dlsprgspa3yayu7kpqkxe

Machine Learning Force Fields [article]

Oliver T. Unke, Stefan Chmiela, Huziel E. Sauceda, Michael Gastegger, Igor Poltavsky, Kristof T. Schütt, Alexandre Tkatchenko, Klaus-Robert Müller
2021 arXiv   pre-print
The key idea is to learn the statistical relation between chemical structure and potential energy without relying on a preconceived notion of fixed chemical bonds or knowledge about the relevant interactions  ...  The core concepts underlying ML-FFs are described in detail and a step-by-step guide for constructing and testing them from scratch is given.  ...  DeePMD-kit The DeePMD-kit is a package written in Python/C++ aiming to minimize the effort required to build deep NNPs with different structural descriptors.  ... 
arXiv:2010.07067v2 fatcat:shqhw36dibf2hgteim3xeywxxu

Isotope Effects in Liquid Water via Deep Potential Molecular Dynamics [article]

Hsin-Yu Ko, Linfeng Zhang, Biswajit Santra, Han Wang, Weinan E, Robert A. DiStasio Jr., Roberto Car
2019 arXiv   pre-print
In particular, we employ the recently developed deep potential molecular dynamics (DPMD) model---a neural-network representation of the ab initio PES---in conjunction with a PI approach based on the generalized  ...  potential energy surface (PES) as well as extensive sampling of configuration space.  ...  Among the suite of existing ML methods, the deep potential molecular dynamics (DPMD) approach [55] [56] [57] utilizes a deep NN to represent the many-body potential energy as a sum of auxiliary "atomic  ... 
arXiv:1904.04930v1 fatcat:isbacyfnnrbjrpcrbnwpu7t3s4

Raman Spectrum and Polarizability of Liquid Water from Deep Neural Networks [article]

Grace M. Sommers, Marcos F. Calegari Andrade, Linfeng Zhang, Han Wang, Roberto Car
2020 arXiv   pre-print
In combination with a neural network representation of the interatomic potential energy surface,the scheme allows us to calculate the Raman spectra along 2-nanosecond classical trajectories at different  ...  We introduce a scheme based on machine learning and deep neural networks to model the environmental dependence of the electronic polarizability in insulating materials.  ...  We used resources of the National Energy Research Scientific Computing Center (DoE Contract No. DE-AC02-05cH11231).  ... 
arXiv:2004.07369v1 fatcat:c73gv2jy2nbvnkuqn74wesfuly

Integrating Machine Learning with Physics-Based Modeling [article]

Weinan E, Jiequn Han, Linfeng Zhang
2020 arXiv   pre-print
Molecular dynamics and moment closure of kinetic equations are used as examples to illustrate the main issues discussed.  ...  Machine learning is poised as a very powerful tool that can drastically improve our ability to carry out scientific research. However, many issues need to be addressed before this becomes a reality.  ...  Deep Potential and DP-GEN have been implemented in the open-source packages DeePMD-kit [29] and DP-GEN [30] , respectively, and have attracted researchers from various disciplines.  ... 
arXiv:2006.02619v1 fatcat:dbaswddd7bdljjhl3can7tjzfm
« Previous Showing results 1 — 15 out of 32 results