Metraipolis › archive › dc_c45bc008be61

Too Rare to Learn: Prescribed Cyclone Tracks Degrade a Bay of Bengal Ocean Emulator — arXiv:2609.04635

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.04635
License (as archived)
CC0-1.0
Archived
2026-10-16T22:57:58.864Z
SHA-256 of the archived text
60fd640c24f17ad3152b44cafda8149efe1e927ab153cee266662e1169ff5678
Stable URL of this page
https://metraipolis.com/documento/dc_c45bc008be61

How to cite

Too Rare to Learn: Prescribed Cyclone Tracks Degrade a Bay of Bengal Ocean Emulator — arXiv:2609.04635 — arXiv (en), https://arxiv.org/abs/2609.04635 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c45bc008be61 (Metraipolis archive dc_c45bc008be61, sha256 60fd640c24f17ad3152b44cafda8149efe1e927ab153cee266662e1169ff5678)
BibTeX
@misc{dcc45bc008be61,
  title = {Too Rare to Learn: Prescribed Cyclone Tracks Degrade a Bay of Bengal Ocean Emulator — arXiv:2609.04635},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.04635},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c45bc008be61; sha256 60fd640c24f17ad3152b44cafda8149efe1e927ab153cee266662e1169ff5678},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.04635. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Neural ocean emulators are being proposed for regional forecasting in cyclone-exposed coastal seas, and a natural design choice is to hand the network the cyclone as a prescribed input. We test that choice in the Bay of Bengal and find it harmful. We withhold 15 whole cyclones spanning 65 to 150 kt from GLORYS12 reanalysis and compare two U-Nets that are identical except for four prescribed cyclone-track channels. Across three seeds the ocean-only model beats persistence in every run and the storm-conditioned model loses to it in every run, with the two skill ranges disjoint (p = 3.1e-5, paired across storms). The cause is exposure frequency rather than signal content: the channels are non-zero on only 7.9% of training days, so they are out of distribution the moment they activate. The extra error falls inside the prescribed storm footprint, and replacing the real cyclone map with a no-storm map at inference improves held-out storm forecasts by 7.5 to 16.4% in every seed. The conditioned network has learned a response to a rare signal that is confidently wrong.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-07 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.04635 — 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 2026-10-16T22:31:32.178Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).