Optimizing transition states via kernel-based machine learning

Zachary D. Pozun, Katja Hansen, Daniel Sheppard, Matthias Rupp, Klaus Muller, Graeme Henkelman

Research output: Contribution to journalArticle

45 Citations (Scopus)

Abstract

We present a method for optimizing transition state theory dividing surfaces with support vector machines. The resulting dividing surfaces require no a priori information or intuition about reaction mechanisms. To generate optimal dividing surfaces, we apply a cycle of machine-learning and refinement of the surface by molecular dynamics sampling. We demonstrate that the machine-learned surfaces contain the relevant low-energy saddle points. The mechanisms of reactions may be extracted from the machine-learned surfaces in order to identify unexpected chemically relevant processes. Furthermore, we show that the machine-learned surfaces significantly increase the transmission coefficient for an adatom exchange involving many coupled degrees of freedom on a (100) surface when compared to a distance-based dividing surface.

Original languageEnglish
Article number174101
JournalJournal of Chemical Physics
Volume136
Issue number17
DOIs
Publication statusPublished - 2012 May 7

Fingerprint

Intuition
machine learning
Molecular Dynamics Simulation
Learning systems
Support Vector Machine
Machine Learning
Adatoms
saddle points
adatoms
Support vector machines
Molecular dynamics
degrees of freedom
sampling
molecular dynamics
Sampling
cycles

ASJC Scopus subject areas

  • Physics and Astronomy(all)
  • Physical and Theoretical Chemistry
  • Medicine(all)

Cite this

Pozun, Z. D., Hansen, K., Sheppard, D., Rupp, M., Muller, K., & Henkelman, G. (2012). Optimizing transition states via kernel-based machine learning. Journal of Chemical Physics, 136(17), [174101]. https://doi.org/10.1063/1.4707167

Optimizing transition states via kernel-based machine learning. / Pozun, Zachary D.; Hansen, Katja; Sheppard, Daniel; Rupp, Matthias; Muller, Klaus; Henkelman, Graeme.

In: Journal of Chemical Physics, Vol. 136, No. 17, 174101, 07.05.2012.

Research output: Contribution to journalArticle

Pozun, ZD, Hansen, K, Sheppard, D, Rupp, M, Muller, K & Henkelman, G 2012, 'Optimizing transition states via kernel-based machine learning', Journal of Chemical Physics, vol. 136, no. 17, 174101. https://doi.org/10.1063/1.4707167
Pozun ZD, Hansen K, Sheppard D, Rupp M, Muller K, Henkelman G. Optimizing transition states via kernel-based machine learning. Journal of Chemical Physics. 2012 May 7;136(17). 174101. https://doi.org/10.1063/1.4707167
Pozun, Zachary D. ; Hansen, Katja ; Sheppard, Daniel ; Rupp, Matthias ; Muller, Klaus ; Henkelman, Graeme. / Optimizing transition states via kernel-based machine learning. In: Journal of Chemical Physics. 2012 ; Vol. 136, No. 17.
@article{809fcdb9a2b84721964940bef534c4cb,
title = "Optimizing transition states via kernel-based machine learning",
abstract = "We present a method for optimizing transition state theory dividing surfaces with support vector machines. The resulting dividing surfaces require no a priori information or intuition about reaction mechanisms. To generate optimal dividing surfaces, we apply a cycle of machine-learning and refinement of the surface by molecular dynamics sampling. We demonstrate that the machine-learned surfaces contain the relevant low-energy saddle points. The mechanisms of reactions may be extracted from the machine-learned surfaces in order to identify unexpected chemically relevant processes. Furthermore, we show that the machine-learned surfaces significantly increase the transmission coefficient for an adatom exchange involving many coupled degrees of freedom on a (100) surface when compared to a distance-based dividing surface.",
author = "Pozun, {Zachary D.} and Katja Hansen and Daniel Sheppard and Matthias Rupp and Klaus Muller and Graeme Henkelman",
year = "2012",
month = "5",
day = "7",
doi = "10.1063/1.4707167",
language = "English",
volume = "136",
journal = "Journal of Chemical Physics",
issn = "0021-9606",
publisher = "American Institute of Physics Publising LLC",
number = "17",

}

TY - JOUR

T1 - Optimizing transition states via kernel-based machine learning

AU - Pozun, Zachary D.

AU - Hansen, Katja

AU - Sheppard, Daniel

AU - Rupp, Matthias

AU - Muller, Klaus

AU - Henkelman, Graeme

PY - 2012/5/7

Y1 - 2012/5/7

N2 - We present a method for optimizing transition state theory dividing surfaces with support vector machines. The resulting dividing surfaces require no a priori information or intuition about reaction mechanisms. To generate optimal dividing surfaces, we apply a cycle of machine-learning and refinement of the surface by molecular dynamics sampling. We demonstrate that the machine-learned surfaces contain the relevant low-energy saddle points. The mechanisms of reactions may be extracted from the machine-learned surfaces in order to identify unexpected chemically relevant processes. Furthermore, we show that the machine-learned surfaces significantly increase the transmission coefficient for an adatom exchange involving many coupled degrees of freedom on a (100) surface when compared to a distance-based dividing surface.

AB - We present a method for optimizing transition state theory dividing surfaces with support vector machines. The resulting dividing surfaces require no a priori information or intuition about reaction mechanisms. To generate optimal dividing surfaces, we apply a cycle of machine-learning and refinement of the surface by molecular dynamics sampling. We demonstrate that the machine-learned surfaces contain the relevant low-energy saddle points. The mechanisms of reactions may be extracted from the machine-learned surfaces in order to identify unexpected chemically relevant processes. Furthermore, we show that the machine-learned surfaces significantly increase the transmission coefficient for an adatom exchange involving many coupled degrees of freedom on a (100) surface when compared to a distance-based dividing surface.

UR - http://www.scopus.com/inward/record.url?scp=84862891798&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84862891798&partnerID=8YFLogxK

U2 - 10.1063/1.4707167

DO - 10.1063/1.4707167

M3 - Article

C2 - 22583204

AN - SCOPUS:84862891798

VL - 136

JO - Journal of Chemical Physics

JF - Journal of Chemical Physics

SN - 0021-9606

IS - 17

M1 - 174101

ER -