Classification of conductance traces with recurrent neural networks

Kasper P. Lauritzen, András Magyarkuti, Zoltán Balogh, A. Halbritter, Gemma C. Solomon

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

We present a new automated method for structural classification of the traces obtained in break junction experiments. Using recurrent neural networks trained on the traces of minimal cross-sectional area in molecular dynamics simulations, we successfully separate the traces into two classes: point contact or nanowire. This is done without any assumptions about the expected features of each class. The trained neural network is applied to experimental break junction conductance traces, and it separates the classes as well as the previously used experimental methods. The effect of using partial conductance traces is explored, and we show that the method performs equally well using full or partial traces (as long as the trace just prior to breaking is included). When only the initial part of the trace is included, the results are still better than random chance. Finally, we show that the neural network classification method can be used to classify experimental conductance traces without using simulated results for training, but instead training the network on a few representative experimental traces. This offers a tool to recognize some characteristic motifs of the traces, which can be hard to find by simple data selection algorithms.

Original languageEnglish
Article number084111
JournalJournal of Chemical Physics
Volume148
Issue number8
DOIs
Publication statusPublished - Feb 28 2018

Fingerprint

Recurrent neural networks
Neural networks
Point contacts
Nanowires
Molecular dynamics
education
Computer simulation
nanowires
Experiments
molecular dynamics
simulation

ASJC Scopus subject areas

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

Cite this

Classification of conductance traces with recurrent neural networks. / Lauritzen, Kasper P.; Magyarkuti, András; Balogh, Zoltán; Halbritter, A.; Solomon, Gemma C.

In: Journal of Chemical Physics, Vol. 148, No. 8, 084111, 28.02.2018.

Research output: Contribution to journalArticle

Lauritzen, Kasper P. ; Magyarkuti, András ; Balogh, Zoltán ; Halbritter, A. ; Solomon, Gemma C. / Classification of conductance traces with recurrent neural networks. In: Journal of Chemical Physics. 2018 ; Vol. 148, No. 8.
@article{119df5c8a7f14448bbf405e9dd343dad,
title = "Classification of conductance traces with recurrent neural networks",
abstract = "We present a new automated method for structural classification of the traces obtained in break junction experiments. Using recurrent neural networks trained on the traces of minimal cross-sectional area in molecular dynamics simulations, we successfully separate the traces into two classes: point contact or nanowire. This is done without any assumptions about the expected features of each class. The trained neural network is applied to experimental break junction conductance traces, and it separates the classes as well as the previously used experimental methods. The effect of using partial conductance traces is explored, and we show that the method performs equally well using full or partial traces (as long as the trace just prior to breaking is included). When only the initial part of the trace is included, the results are still better than random chance. Finally, we show that the neural network classification method can be used to classify experimental conductance traces without using simulated results for training, but instead training the network on a few representative experimental traces. This offers a tool to recognize some characteristic motifs of the traces, which can be hard to find by simple data selection algorithms.",
author = "Lauritzen, {Kasper P.} and Andr{\'a}s Magyarkuti and Zolt{\'a}n Balogh and A. Halbritter and Solomon, {Gemma C.}",
year = "2018",
month = "2",
day = "28",
doi = "10.1063/1.5012514",
language = "English",
volume = "148",
journal = "Journal of Chemical Physics",
issn = "0021-9606",
publisher = "American Institute of Physics Publising LLC",
number = "8",

}

TY - JOUR

T1 - Classification of conductance traces with recurrent neural networks

AU - Lauritzen, Kasper P.

AU - Magyarkuti, András

AU - Balogh, Zoltán

AU - Halbritter, A.

AU - Solomon, Gemma C.

PY - 2018/2/28

Y1 - 2018/2/28

N2 - We present a new automated method for structural classification of the traces obtained in break junction experiments. Using recurrent neural networks trained on the traces of minimal cross-sectional area in molecular dynamics simulations, we successfully separate the traces into two classes: point contact or nanowire. This is done without any assumptions about the expected features of each class. The trained neural network is applied to experimental break junction conductance traces, and it separates the classes as well as the previously used experimental methods. The effect of using partial conductance traces is explored, and we show that the method performs equally well using full or partial traces (as long as the trace just prior to breaking is included). When only the initial part of the trace is included, the results are still better than random chance. Finally, we show that the neural network classification method can be used to classify experimental conductance traces without using simulated results for training, but instead training the network on a few representative experimental traces. This offers a tool to recognize some characteristic motifs of the traces, which can be hard to find by simple data selection algorithms.

AB - We present a new automated method for structural classification of the traces obtained in break junction experiments. Using recurrent neural networks trained on the traces of minimal cross-sectional area in molecular dynamics simulations, we successfully separate the traces into two classes: point contact or nanowire. This is done without any assumptions about the expected features of each class. The trained neural network is applied to experimental break junction conductance traces, and it separates the classes as well as the previously used experimental methods. The effect of using partial conductance traces is explored, and we show that the method performs equally well using full or partial traces (as long as the trace just prior to breaking is included). When only the initial part of the trace is included, the results are still better than random chance. Finally, we show that the neural network classification method can be used to classify experimental conductance traces without using simulated results for training, but instead training the network on a few representative experimental traces. This offers a tool to recognize some characteristic motifs of the traces, which can be hard to find by simple data selection algorithms.

UR - http://www.scopus.com/inward/record.url?scp=85042734095&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85042734095&partnerID=8YFLogxK

U2 - 10.1063/1.5012514

DO - 10.1063/1.5012514

M3 - Article

VL - 148

JO - Journal of Chemical Physics

JF - Journal of Chemical Physics

SN - 0021-9606

IS - 8

M1 - 084111

ER -