Testing the reliability of interpretable neural networks in geoscience using the Madden–Julian oscillation

Toms, Benjamin A.; Kashinath, Karthik; Prabhat,; Yang, Da

doi:https://doi.org/10.5194/gmd-14-4495-2021

Articles | Volume 14, issue 7

https://doi.org/10.5194/gmd-14-4495-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

https://doi.org/10.5194/gmd-14-4495-2021

© Author(s) 2021. This work is distributed under
the Creative Commons Attribution 4.0 License.

Articles | Volume 14, issue 7

Methods for assessment of models

|

22 Jul 2021

Methods for assessment of models |

| 22 Jul 2021

Testing the reliability of interpretable neural networks in geoscience using the Madden–Julian oscillation

Benjamin A. Toms, Karthik Kashinath, Prabhat, and Da Yang

Download

Final revised paper (published on 22 Jul 2021)
Preprint (discussion started on 03 Aug 2020)

Interactive discussion

Status: closed

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

- Printer-friendly version

- Supplement

RC1: 'Review of "Testing the Reliability of Interpretable Neural Networks in Geoscience Using the Madden-Julian Oscillation"', Anonymous Referee #1, 10 Aug 2020
- AC1: 'Authors' Response to Reviewer 1', Benjamin Toms, 18 Aug 2020
RC2: 'Review of GMD-2020-152', Anonymous Referee #2, 04 Sep 2020
- AC2: 'Authors' Response to Reviewer 2', Benjamin Toms, 21 Dec 2020

Peer-review completion

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

AR by Benjamin Toms on behalf of the Authors (21 Dec 2020) Manuscript

ED: Publish as is (13 Feb 2021) by Richard Neale

AR by Benjamin Toms on behalf of the Authors (20 Mar 2021)

Short summary

We test whether a type of machine learning called neural networks can be used trustfully within the geosciences. We do so by challenging the networks to understand the spatial patterns of a commonly studied geoscientific phenomenon. The neural networks can correctly identify the spatial patterns, which lends confidence that similar networks can be used for more uncertain problems. The results of this study may give geoscientists confidence when using neural networks in their research.