Stochastic Neural Network Approach for Learning High-Dimensional Free Energy Surfaces

Elia Schneider, Luke Dai, Robert Q. Topper, Christof Drechsel-Grau, Mark Tuckerman

Research output: Contribution to journalArticle

Abstract

The generation of free energy landscapes corresponding to conformational equilibria in complex molecular systems remains a significant computational challenge. Adding to this challenge is the need to represent, store, and manipulate the often high-dimensional surfaces that result from rare-event sampling approaches employed to compute them. In this Letter, we propose the use of artificial neural networks as a solution to these issues. Using specific examples, we discuss network training using enhanced-sampling methods and the use of the networks in the calculation of ensemble averages.

Original languageEnglish (US)
Article number150601
JournalPhysical Review Letters
Volume119
Issue number15
DOIs
StatePublished - Oct 11 2017

Fingerprint

learning
free energy
sampling
education

ASJC Scopus subject areas

  • Physics and Astronomy(all)

Cite this

Stochastic Neural Network Approach for Learning High-Dimensional Free Energy Surfaces. / Schneider, Elia; Dai, Luke; Topper, Robert Q.; Drechsel-Grau, Christof; Tuckerman, Mark.

In: Physical Review Letters, Vol. 119, No. 15, 150601, 11.10.2017.

Research output: Contribution to journalArticle

Schneider, Elia ; Dai, Luke ; Topper, Robert Q. ; Drechsel-Grau, Christof ; Tuckerman, Mark. / Stochastic Neural Network Approach for Learning High-Dimensional Free Energy Surfaces. In: Physical Review Letters. 2017 ; Vol. 119, No. 15.
@article{e332ffd2c3ec4a629e4a40b6dafcd1d5,
title = "Stochastic Neural Network Approach for Learning High-Dimensional Free Energy Surfaces",
abstract = "The generation of free energy landscapes corresponding to conformational equilibria in complex molecular systems remains a significant computational challenge. Adding to this challenge is the need to represent, store, and manipulate the often high-dimensional surfaces that result from rare-event sampling approaches employed to compute them. In this Letter, we propose the use of artificial neural networks as a solution to these issues. Using specific examples, we discuss network training using enhanced-sampling methods and the use of the networks in the calculation of ensemble averages.",
author = "Elia Schneider and Luke Dai and Topper, {Robert Q.} and Christof Drechsel-Grau and Mark Tuckerman",
year = "2017",
month = "10",
day = "11",
doi = "10.1103/PhysRevLett.119.150601",
language = "English (US)",
volume = "119",
journal = "Physical Review Letters",
issn = "0031-9007",
publisher = "American Physical Society",
number = "15",

}

TY - JOUR

T1 - Stochastic Neural Network Approach for Learning High-Dimensional Free Energy Surfaces

AU - Schneider, Elia

AU - Dai, Luke

AU - Topper, Robert Q.

AU - Drechsel-Grau, Christof

AU - Tuckerman, Mark

PY - 2017/10/11

Y1 - 2017/10/11

N2 - The generation of free energy landscapes corresponding to conformational equilibria in complex molecular systems remains a significant computational challenge. Adding to this challenge is the need to represent, store, and manipulate the often high-dimensional surfaces that result from rare-event sampling approaches employed to compute them. In this Letter, we propose the use of artificial neural networks as a solution to these issues. Using specific examples, we discuss network training using enhanced-sampling methods and the use of the networks in the calculation of ensemble averages.

AB - The generation of free energy landscapes corresponding to conformational equilibria in complex molecular systems remains a significant computational challenge. Adding to this challenge is the need to represent, store, and manipulate the often high-dimensional surfaces that result from rare-event sampling approaches employed to compute them. In this Letter, we propose the use of artificial neural networks as a solution to these issues. Using specific examples, we discuss network training using enhanced-sampling methods and the use of the networks in the calculation of ensemble averages.

UR - http://www.scopus.com/inward/record.url?scp=85031325693&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85031325693&partnerID=8YFLogxK

U2 - 10.1103/PhysRevLett.119.150601

DO - 10.1103/PhysRevLett.119.150601

M3 - Article

VL - 119

JO - Physical Review Letters

JF - Physical Review Letters

SN - 0031-9007

IS - 15

M1 - 150601

ER -