On-the-Fly Active Learning of Interatomic Potentials for Large-Scale Atomistic Simulations

2020 ◽  
Vol 11 (17) ◽  
pp. 6946-6955 ◽  
Author(s):  
Ryosuke Jinnouchi ◽  
Kazutoshi Miwa ◽  
Ferenc Karsai ◽  
Georg Kresse ◽  
Ryoji Asahi
Keyword(s):  
Active Learning ◽  
Large Scale ◽  
Atomistic Simulations ◽  
Interatomic Potentials

scholarly journals A Transferable Active-Learning Strategy for Reactive Molecular Force Fields

2021 ◽  
Author(s):  
Tom Young ◽  
Tristan Johnston-Wood ◽  
Volker Deringer ◽  
Fernanda Duarte
Keyword(s):  
Active Learning ◽  
Large Scale ◽  
Learning Strategy ◽  
Gaussian Approximation ◽  
Force Fields ◽  
Molecular System ◽  
Chemical Reactivity ◽  
Bulk Water ◽  
Interatomic Potentials ◽  
Molecular Systems

<p>Predictive simulations of dynamic processes in molecular systems require fast, accurate and reactive interatomic potentials. Machine learning offers a promising approach to construct force-field models for large-scale molecular simulation by fitting to high-level quantum-mechanical data. However, machine-learned force fields generally require considerable human intervention and data volume. Here we show that, by leveraging hierarchical and active learning, accurate Gaussian Approximation Potential (GAP) models for diverse chemical systems can be developed in an autonomous way, requiring only hundreds to a few thousand energy and gradient evaluations on the reference potential-energy surface. Our approach relies on a decomposition of the condensed-phase molecular system into intra- and inter-molecular terms, and on the definition of a prospective error metric to quantify accuracy. We demonstrate applications to a range of molecular systems: from bulk water, organic solvents, and a solvated ion onwards to the description of chemical reactivity, including, a bifurcating Diels–Alder reaction in the gas phase and non-equilibrium dynamics (S<sub>N</sub>2 reaction) in explicit solvent. The method provides a route to routinely generating machine-learned force fields for complex and/or reactive molecular systems. </p>

Interatomic Potentials for Atomistic Simulations

MRS Bulletin ◽  
1996 ◽  
Vol 21 (2) ◽  
pp. 17-19 ◽  
Author(s):  
Arthur F. Voter
Keyword(s):  
Atomistic Simulation ◽  
Large Scale ◽  
Materials Science ◽  
Interatomic Potential ◽  
Computer Time ◽  
Atomistic Simulations ◽  
Interatomic Potentials ◽  
Materials Research ◽  
Recent Developments ◽  
Time Required

Atomistic simulations are playing an increasingly prominent role in materials science. From relatively conventional studies of point and planar defects to large-scale simulations of fracture and machining, atomistic simulations offer a microscopic view of the physics that cannot be obtained from experiment. Predictions resulting from this atomic-level understanding are proving increasingly accurate and useful. Consequently, the field of atomistic simulation is gaining ground as an indispensable partner in materials research, a trend that can only continue. Each year, computers gain roughly a factor of two in speed. With the same effort one can then simulate a system with twice as many atoms or integrate a molecular-dynamics trajectory for twice as long. Perhaps even more important, however, are the theoretical advances occurring in the description of the atomic interactions, the so-called “interatomic potential” function.The interatomic potential underpins any atomistic simulation. The accuracy of the potential dictates the quality of the simulation results, and its functional complexity determines the amount of computer time required. Recent developments that fit more physics into a compact potential form are increasing the accuracy available per simulation dollar.This issue of MRS Bulletin offers an introductory survey of interatomic potentials in use today, as well as the types of problems to which they can be applied. This is by no means a comprehensive review. It would be impractical here to attempt to present all the potentials that have been developed in recent years. Rather, this collection of articles focuses on a few important forms of potential spanning the major classes of materials bonding: covalent, metallic, and ionic.

scholarly journals A Transferable Active-Learning Strategy for Reactive Molecular Force Fields

2021 ◽  
Author(s):  
Tom Young ◽  
Tristan Johnston-Wood ◽  
Volker Deringer ◽  
Fernanda Duarte
Keyword(s):  
Active Learning ◽  
Large Scale ◽  
Learning Strategy ◽  
Gaussian Approximation ◽  
Force Fields ◽  
Molecular System ◽  
Chemical Reactivity ◽  
Bulk Water ◽  
Interatomic Potentials ◽  
Molecular Systems

<p>Predictive simulations of dynamic processes in molecular systems require fast, accurate and reactive interatomic potentials. Machine learning offers a promising approach to construct force-field models for large-scale molecular simulation by fitting to high-level quantum-mechanical data. However, machine-learned force fields generally require considerable human intervention and data volume. Here we show that, by leveraging hierarchical and active learning, accurate Gaussian Approximation Potential (GAP) models for diverse chemical systems can be developed in an autonomous way, requiring only hundreds to a few thousand energy and gradient evaluations on the reference potential-energy surface. Our approach relies on a decomposition of the condensed-phase molecular system into intra- and inter-molecular terms, and on the definition of a prospective error metric to quantify accuracy. We demonstrate applications to a range of molecular systems: from bulk water, organic solvents, and a solvated ion onwards to the description of chemical reactivity, including, a bifurcating Diels–Alder reaction in the gas phase and non-equilibrium dynamics (S<sub>N</sub>2 reaction) in explicit solvent. The method provides a route to routinely generating machine-learned force fields for complex and/or reactive molecular systems. </p>

scholarly journals Performant implementation of the atomic cluster expansion (PACE) and application to copper and silicon

2021 ◽  
Vol 7 (1) ◽  
Author(s):  
Yury Lysogorskiy ◽  
Cas van der Oord ◽  
Anton Bochkarev ◽  
Sarath Menon ◽  
Matteo Rinaldi ◽  
...  
Keyword(s):  
Machine Learning ◽  
Cluster Expansion ◽  
Large Scale ◽  
Pareto Front ◽  
General Purpose ◽  
Basis Functions ◽  
Atomic Energy ◽  
Atomistic Simulations ◽  
Interatomic Potentials ◽  
Atomic Cluster

AbstractThe atomic cluster expansion is a general polynomial expansion of the atomic energy in multi-atom basis functions. Here we implement the atomic cluster expansion in the performant C++ code that is suitable for use in large-scale atomistic simulations. We briefly review the atomic cluster expansion and give detailed expressions for energies and forces as well as efficient algorithms for their evaluation. We demonstrate that the atomic cluster expansion as implemented in shifts a previously established Pareto front for machine learning interatomic potentials toward faster and more accurate calculations. Moreover, general purpose parameterizations are presented for copper and silicon and evaluated in detail. We show that the Cu and Si potentials significantly improve on the best available potentials for highly accurate large-scale atomistic simulations.

scholarly journals A Transferable Active-Learning Strategy for Reactive Molecular Force Fields

2021 ◽  
Author(s):  
Tom Young ◽  
Tristan Johnston-Wood ◽  
Volker Deringer ◽  
Fernanda Duarte
Keyword(s):  
Active Learning ◽  
Large Scale ◽  
Learning Strategy ◽  
Gaussian Approximation ◽  
Force Fields ◽  
Molecular System ◽  
Chemical Reactivity ◽  
Bulk Water ◽  
Interatomic Potentials ◽  
Molecular Systems

Predictive simulations of dynamic processes in molecular systems require fast, accurate and reactive interatomic potentials. Machine learning offers a promising approach to construct force-field models for large-scale molecular simulation by fitting to high-level quantum-mechanical data. However, machine-learned force fields generally require considerable human intervention and data volume. Here we show that, by leveraging hierarchical and active learning, accurate Gaussian Approximation Potential (GAP) models for diverse chemical systems can be developed in an autonomous way, requiring only hundreds to a few thousand energy and gradient evaluations on the reference potential-energy surface. Our approach relies on a decomposition of the condensed-phase molecular system into intra- and inter-molecular terms, and on the definition of a prospective error metric to quantify accuracy. We demonstrate applications to a range of molecular systems: from bulk water, organic solvents, and a solvated ion onwards to the description of chemical reactivity, including, a bifurcating Diels–Alder reaction in the gas phase and non-equilibrium dynamics (SN2 reaction) in explicit solvent. The method provides a route to routinely generating machine-learned force fields for complex and/or reactive molecular systems.

Reaction-based Enumeration, Active Learning, and Free Energy Calculations to Rapidly Explore Synthetically Tractable Chemical Space and Optimize Potency of Cyclin Dependent Kinase 2 Inhibitors

2019 ◽  
Author(s):  
Kyle Konze ◽  
Pieter Bos ◽  
Markus Dahlgren ◽  
Karl Leswing ◽  
Ivan Tubert-Brohman ◽  
...  
Keyword(s):  
Free Energy ◽  
Drug Discovery ◽  
Active Learning ◽  
Large Scale ◽  
Chemical Space ◽  
Population Based ◽  
Free Energy Calculations ◽  
Computational Technique ◽  
Cyclin Dependent Kinase ◽  
Energy Calculations

We report a new computational technique, PathFinder, that uses retrosynthetic analysis followed by combinatorial synthesis to generate novel compounds in synthetically accessible chemical space. Coupling PathFinder with active learning and cloud-based free energy calculations allows for large-scale potency predictions of compounds on a timescale that impacts drug discovery. The process is further accelerated by using a combination of population-based statistics and active learning techniques. Using this approach, we rapidly optimized R-groups and core hops for inhibitors of cyclin-dependent kinase 2. We explored greater than 300 thousand ideas and identified 35 ligands with diverse commercially available R-groups and a predicted IC<sub>50</sub> < 100 nM, and four unique cores with a predicted IC<sub>50</sub> < 100 nM. The rapid turnaround time, and scale of chemical exploration, suggests that this is a useful approach to accelerate the discovery of novel chemical matter in drug discovery campaigns.

Reaction-based Enumeration, Active Learning, and Free Energy Calculations to Rapidly Explore Synthetically Tractable Chemical Space and Optimize Potency of Cyclin Dependent Kinase 2 Inhibitors

2019 ◽  
Author(s):  
Kyle Konze ◽  
Pieter Bos ◽  
Markus Dahlgren ◽  
Karl Leswing ◽  
Ivan Tubert-Brohman ◽  
...  
Keyword(s):  
Free Energy ◽  
Drug Discovery ◽  
Active Learning ◽  
Large Scale ◽  
Chemical Space ◽  
Population Based ◽  
Free Energy Calculations ◽  
Computational Technique ◽  
Cyclin Dependent Kinase ◽  
Energy Calculations

We report a new computational technique, PathFinder, that uses retrosynthetic analysis followed by combinatorial synthesis to generate novel compounds in synthetically accessible chemical space. Coupling PathFinder with active learning and cloud-based free energy calculations allows for large-scale potency predictions of compounds on a timescale that impacts drug discovery. The process is further accelerated by using a combination of population-based statistics and active learning techniques. Using this approach, we rapidly optimized R-groups and core hops for inhibitors of cyclin-dependent kinase 2. We explored greater than 300 thousand ideas and identified 35 ligands with diverse commercially available R-groups and a predicted IC<sub>50</sub> < 100 nM, and four unique cores with a predicted IC<sub>50</sub> < 100 nM. The rapid turnaround time, and scale of chemical exploration, suggests that this is a useful approach to accelerate the discovery of novel chemical matter in drug discovery campaigns.

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...

Sign in / Sign up

Export Citation Format

Share Document