Neural networks for computational chemistry: Pitfalls and recommendations

Grégoire Montavon, Klaus Muller

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

There is a long history of using neural networks for function approximation in computational physics and chemistry. Despite their conceptual simplicity, the practitioner may face difficulties when it comes to putting them to work. This small guide intends to pinpoint some neural networks pitfalls, along with corresponding solutions to successfully realize function approximation tasks in physics, chemistry or other fields.

Original languageEnglish
Title of host publicationMaterials Research Society Symposium Proceedings
PublisherMaterials Research Society
Pages24-29
Number of pages6
Volume1523
ISBN (Print)9781632661135
DOIs
Publication statusPublished - 2013 Jan 1
Event2012 MRS Fall Meeting - Boston, MA, United States
Duration: 2012 Nov 252012 Nov 30

Other

Other2012 MRS Fall Meeting
CountryUnited States
CityBoston, MA
Period12/11/2512/11/30

Fingerprint

Computational chemistry
computational chemistry
recommendations
Physics
chemistry
Neural networks
physics
approximation
histories

ASJC Scopus subject areas

  • Materials Science(all)
  • Condensed Matter Physics
  • Mechanical Engineering
  • Mechanics of Materials

Cite this

Montavon, G., & Muller, K. (2013). Neural networks for computational chemistry: Pitfalls and recommendations. In Materials Research Society Symposium Proceedings (Vol. 1523, pp. 24-29). Materials Research Society. https://doi.org/10.1557/opl.2013.189

Neural networks for computational chemistry : Pitfalls and recommendations. / Montavon, Grégoire; Muller, Klaus.

Materials Research Society Symposium Proceedings. Vol. 1523 Materials Research Society, 2013. p. 24-29.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Montavon, G & Muller, K 2013, Neural networks for computational chemistry: Pitfalls and recommendations. in Materials Research Society Symposium Proceedings. vol. 1523, Materials Research Society, pp. 24-29, 2012 MRS Fall Meeting, Boston, MA, United States, 12/11/25. https://doi.org/10.1557/opl.2013.189
Montavon G, Muller K. Neural networks for computational chemistry: Pitfalls and recommendations. In Materials Research Society Symposium Proceedings. Vol. 1523. Materials Research Society. 2013. p. 24-29 https://doi.org/10.1557/opl.2013.189
Montavon, Grégoire ; Muller, Klaus. / Neural networks for computational chemistry : Pitfalls and recommendations. Materials Research Society Symposium Proceedings. Vol. 1523 Materials Research Society, 2013. pp. 24-29
@inproceedings{b0d34c1049214a78b379a80afb155054,
title = "Neural networks for computational chemistry: Pitfalls and recommendations",
abstract = "There is a long history of using neural networks for function approximation in computational physics and chemistry. Despite their conceptual simplicity, the practitioner may face difficulties when it comes to putting them to work. This small guide intends to pinpoint some neural networks pitfalls, along with corresponding solutions to successfully realize function approximation tasks in physics, chemistry or other fields.",
author = "Gr{\'e}goire Montavon and Klaus Muller",
year = "2013",
month = "1",
day = "1",
doi = "10.1557/opl.2013.189",
language = "English",
isbn = "9781632661135",
volume = "1523",
pages = "24--29",
booktitle = "Materials Research Society Symposium Proceedings",
publisher = "Materials Research Society",

}

TY - GEN

T1 - Neural networks for computational chemistry

T2 - Pitfalls and recommendations

AU - Montavon, Grégoire

AU - Muller, Klaus

PY - 2013/1/1

Y1 - 2013/1/1

N2 - There is a long history of using neural networks for function approximation in computational physics and chemistry. Despite their conceptual simplicity, the practitioner may face difficulties when it comes to putting them to work. This small guide intends to pinpoint some neural networks pitfalls, along with corresponding solutions to successfully realize function approximation tasks in physics, chemistry or other fields.

AB - There is a long history of using neural networks for function approximation in computational physics and chemistry. Despite their conceptual simplicity, the practitioner may face difficulties when it comes to putting them to work. This small guide intends to pinpoint some neural networks pitfalls, along with corresponding solutions to successfully realize function approximation tasks in physics, chemistry or other fields.

UR - http://www.scopus.com/inward/record.url?scp=84900309094&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84900309094&partnerID=8YFLogxK

U2 - 10.1557/opl.2013.189

DO - 10.1557/opl.2013.189

M3 - Conference contribution

AN - SCOPUS:84900309094

SN - 9781632661135

VL - 1523

SP - 24

EP - 29

BT - Materials Research Society Symposium Proceedings

PB - Materials Research Society

ER -