Metraipolis › archive › dc_c005a251fd28

SoftSEEPS improves ML-based precipitation forecasting — arXiv:2610.09752

Source
arXiv (en)
Original
https://arxiv.org/abs/2610.09752
License (as archived)
CC0-1.0
Archived
2026-10-08T11:59:06.594Z
SHA-256 of the archived text
ecb5eb5ddd699d85a5e708db4ea2fb03f6919e150865eb8ed07c5618187b14a1
Stable URL of this page
https://metraipolis.com/documento/dc_c005a251fd28

How to cite

SoftSEEPS improves ML-based precipitation forecasting — arXiv:2610.09752 — arXiv (en), https://arxiv.org/abs/2610.09752 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c005a251fd28 (Metraipolis archive dc_c005a251fd28, sha256 ecb5eb5ddd699d85a5e708db4ea2fb03f6919e150865eb8ed07c5618187b14a1)
BibTeX
@misc{dcc005a251fd28,
  title = {SoftSEEPS improves ML-based precipitation forecasting — arXiv:2610.09752},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2610.09752},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c005a251fd28; sha256 ecb5eb5ddd699d85a5e708db4ea2fb03f6919e150865eb8ed07c5618187b14a1},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.09752. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

In this paper we have developed a differentiable approximation of the well-known SEEPS score, which we name SoftSEEPS. This allows the training of a Machine Learning model to forecast precipitation directly. We test SoftSEEPS on the IMERG dataset (0.1 degree resolution) by training a decoder for precipitation on the latent space of a pre-trained low-resolution forecasting model. Combining SoftSEEPS and RMSE in a joint objective is possible with marginal trade-offs in either metric.

Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.LG, nuovi lavori del giorno).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2610.09752 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-08T10:50:37.710Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).