Skip to content
DraconiticResearch
Search Sources API
197,768 studies31 sourcesUpdated 48 min ago
Peer-reviewedLicence not statedArtificial Intelligence for the Earth Systems

FastNet: Improving the Physical Consistency of Machine Learning Weather Prediction Models through Loss Function Design

  • Tom Dunstan1
  • Oliver Strickson2
  • Thusal Bennett1
  • Jack Bowyer1
  • Matthew Burnand1
  • James Chappell1
  • Alejandro Coca-Castro2
  • Kirstine Ida Dale1
  • E. G. Daub2
  • Noushin Eftekhari2
  • Manvendra Janmaijaya2
  • Jon Lillis1
25 more authors
  • David Salvador-Jasin2
  • Nathan Daniel Simpson2
  • Ryan Sze-Yin Chan2
  • Mohamad Elmasri3,2
  • Lydia A. France2
  • Sam Madge1
  • Sophie Luise Arana2
  • Levan Bokeria2
  • H. Alex Brown1,4
  • Evangeline Corcoran2
  • Tom Dodds1,5
  • Anna-Louise Ellis1
  • Tomas Lazauskas2
  • David Llewellyn-Jones2
  • Theo McCaie1
  • Sophia Moreton1
  • Tom Potter1
  • James Robinson2
  • Adam A. Scaife6,7
  • Iain Stenson2
  • David Walters1
  • Karina Williams8,1
  • Louisa van Zeeland2
  • Peter Yatsyshin2
  • J. Scott Hosking9,2
9 affiliations
  1. 1Met Office, Exeter, United Kingdom
  2. 2The Alan Turing Institute, London, United Kingdom
  3. 3Carleton University, Ottawa, Ontario, Canada
  4. 4UK Hydrographic Office, Taunton, United Kingdom
  5. 5Office for National Statistics, Newport, United Kingdom
  6. 6Department of Mathematics and Statistics, University of Exeter, Exeter, United Kingdom
  7. 7Met Office Hadley Centre, Exeter, United Kingdom
  8. 8Global Systems Institute, University of Exeter, Exeter, United Kingdom
  9. 9British Antarctic Survey, Cambridge, United Kingdom
Open at sourcePDFDOICite

Related

  1. PhysMetrics.Weather: An Evaluation Framework for Physical Consistency in ML Weather Models
    Emma Kasteleyn, Timo Maier, Axel Lauer +3
    arXiv (Cornell University)PreprintLicence not statedFull text6 citedAtmospheric science & meteorologyOpenAlex
  2. Explainable AI with dual-layer validation framework for ensuring noise resilience and physical consistency in low-cost sustainable water quality monitoring
    Wibowo Harry Sugiharto, Rizkysari Meimaharani, Ahmad Abdul Chamid +2
    Discover SustainabilityPeer-reviewedCC BYFull textHydrology & soilsDOAJ
  3. Biogeochemistry-Informed Neural Network (BINN v1.0) for improving accuracy of model prediction and scientific understanding of soil organic carbon storage
    Haodi Xu, Joshua Fan, Feng Tao +6
    Geoscientific model developmentPeer-reviewedCC BYFull text10 citedHydrology & soilsOpenAlexDOAJ
  4. Improving subsurface characterization in tropical granitic terrains using machine learning and deep learning-enhanced resistivity–chargeability modeling
    Adedibu Sunny Akingboye, Andy Anderson Bery, Hui Tang +4
    Scientific ReportsPeer-reviewedCC BYFull text5 citedGeophysics & seismologyOpenAlex
  5. Improving flood forecasts: the combined impact of data assimilation and machine learning post-processing
    Gustavo Gabbardo dos Reis, Paul Astagneau, François Bourgin +2
    Journal of HydrologyPeer-reviewedCC BYFull textHydrology & soilsHAL

Cite

BibTeXRIS
BibTeX
@article{dunstan2026fastnet,
  title = {{FastNet: Improving the Physical Consistency of Machine Learning Weather Prediction Models through Loss Function Design}},
  author = {Tom Dunstan and Oliver Strickson and Thusal Bennett and Jack Bowyer and Matthew Burnand and James Chappell and Alejandro Coca-Castro and Kirstine Ida Dale and E. G. Daub and Noushin Eftekhari and Manvendra Janmaijaya and Jon Lillis and David Salvador-Jasin and Nathan Daniel Simpson and Ryan Sze-Yin Chan and Mohamad Elmasri and Lydia A. France and Sam Madge and Sophie Luise Arana and Levan Bokeria and H. Alex Brown and Evangeline Corcoran and Tom Dodds and Anna-Louise Ellis and Tomas Lazauskas and David Llewellyn-Jones and Theo McCaie and Sophia Moreton and Tom Potter and James Robinson and Adam A. Scaife and Iain Stenson and David Walters and Karina Williams and Louisa van Zeeland and Peter Yatsyshin and J. Scott Hosking},
  journal = {Artificial Intelligence for the Earth Systems},
  year = {2026},
  month = jul,
  doi = {10.1175/aies-d-25-0090.1},
  url = {https://arxiv.org/abs/2509.17601},
}
RIS
TY  - JOUR
TI  - FastNet: Improving the Physical Consistency of Machine Learning Weather Prediction Models through Loss Function Design
AU  - Tom Dunstan
AU  - Oliver Strickson
AU  - Thusal Bennett
AU  - Jack Bowyer
AU  - Matthew Burnand
AU  - James Chappell
AU  - Alejandro Coca-Castro
AU  - Kirstine Ida Dale
AU  - E. G. Daub
AU  - Noushin Eftekhari
AU  - Manvendra Janmaijaya
AU  - Jon Lillis
AU  - David Salvador-Jasin
AU  - Nathan Daniel Simpson
AU  - Ryan Sze-Yin Chan
AU  - Mohamad Elmasri
AU  - Lydia A. France
AU  - Sam Madge
AU  - Sophie Luise Arana
AU  - Levan Bokeria
AU  - H. Alex Brown
AU  - Evangeline Corcoran
AU  - Tom Dodds
AU  - Anna-Louise Ellis
AU  - Tomas Lazauskas
AU  - David Llewellyn-Jones
AU  - Theo McCaie
AU  - Sophia Moreton
AU  - Tom Potter
AU  - James Robinson
AU  - Adam A. Scaife
AU  - Iain Stenson
AU  - David Walters
AU  - Karina Williams
AU  - Louisa van Zeeland
AU  - Peter Yatsyshin
AU  - J. Scott Hosking
PY  - 2026
DA  - 2026/07/01
JO  - Artificial Intelligence for the Earth Systems
DO  - 10.1175/aies-d-25-0090.1
UR  - https://arxiv.org/abs/2509.17601
SN  - 2769-7525
ER  -