Generalization properties of feed-forward neural networks trained on Lorenz systems

Scher, Sebastian; Messori, Gabriele

doi:https://doi.org/10.5194/npg-26-381-2019

Articles | Volume 26, issue 4

https://doi.org/10.5194/npg-26-381-2019

© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/npg-26-381-2019

© Author(s) 2019. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 26, issue 4

Research article

|

05 Nov 2019

Research article |

| 05 Nov 2019

Generalization properties of feed-forward neural networks trained on Lorenz systems

Sebastian Scher and Gabriele Messori

Download

Final revised paper (published on 05 Nov 2019)
Preprint (discussion started on 14 Jun 2019)

Interactive discussion

Status: closed

AC: Author comment | RC: Referee comment | SC: Short comment | EC: Editor comment

- Printer-friendly version

- Supplement

RC1: 'Review', Anonymous Referee #1, 24 Jun 2019
- SC1: 'clarifications', Sebastian Scher, 26 Jun 2019
RC2: 'Review', Anonymous Referee #2, 25 Jul 2019
RC3: 'Referee report', Anonymous Referee #3, 31 Jul 2019
AC1: 'final-response', Sebastian Scher, 12 Aug 2019

Peer-review completion

AR: Author's response | RR: Referee report | ED: Editor decision

AR by Sebastian Scher on behalf of the Authors (26 Sep 2019) Author's response Manuscript

ED: Publish as is (06 Oct 2019) by Stefano Pierini

AR by Sebastian Scher on behalf of the Authors (08 Oct 2019) Manuscript

Short summary

Neural networks are a technique that is widely used to predict the time evolution of physical systems. For this the past evolution of the system is shown to the neural network – it is trained – and then can be used to predict the evolution in the future. We show some limitations in this approach for certain systems that are important to consider when using neural networks for climate- and weather-related applications.