Metraipolis › archive › dc_a2b4e19b80c8

OceanMoE: Structured Conditional Sparse Computation for Long-Horizon Multivariate Ocean Forecasting — arXiv:2609.19768

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.19768
License (as archived)
CC0-1.0
Archived
2026-10-10T04:08:43.859Z
SHA-256 of the archived text
9160da957a1d6052b3b7af2c1c22c1fda7d0c7786d5c41f651e7438c3b9d1f01
Stable URL of this page
https://metraipolis.com/documento/dc_a2b4e19b80c8

How to cite

OceanMoE: Structured Conditional Sparse Computation for Long-Horizon Multivariate Ocean Forecasting — arXiv:2609.19768 — arXiv (en), https://arxiv.org/abs/2609.19768 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a2b4e19b80c8 (Metraipolis archive dc_a2b4e19b80c8, sha256 9160da957a1d6052b3b7af2c1c22c1fda7d0c7786d5c41f651e7438c3b9d1f01)
BibTeX
@misc{dca2b4e19b80c8,
  title = {OceanMoE: Structured Conditional Sparse Computation for Long-Horizon Multivariate Ocean Forecasting — arXiv:2609.19768},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.19768},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a2b4e19b80c8; sha256 9160da957a1d6052b3b7af2c1c22c1fda7d0c7786d5c41f651e7438c3b9d1f01},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.19768. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multivariate ocean forecasting must exploit shared evolution in a coupled ocean system while adapting to the heterogeneous statistical and dynamical characteristics of different prediction variables and locations. Fully shared models may lack the flexibility to handle this heterogeneity, whereas fully independent models discard the common ocean context shared across variables. The key question is how to retain shared context in a unified model while allowing computation to specialize according to the prediction target and local state. We propose OceanMoE, a structured conditional sparse Mixture-of-Experts framework that combines sharing and specialization for multivariate ocean forecasting. OceanMoE fuses cross-variable information to construct target-specific local representations and uses them to perform content-conditioned sparse routing at each spatial location, with the number of active experts adapted to router confidence. In the decoder, routing is augmented with a learned geographic bias parameterized by spherical-harmonic spatial bases, while shared residual and seasonal pathways provide common cross-variable and month-dependent context. Experiments on long-horizon autoregressive ORAS5 forecasting show that OceanMoE lowers aggregate forecasting error in both evaluated settings and maintains lower geometric-mean normalized RMSE than the corresponding baselines over most later rollout months. Routing analyses further show that expert allocation varies with prediction targets and spatial locations. These results support structured conditional computation as a modeling strategy for balancing shared ocean context with adaptive specialization.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-18 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.19768 — 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 2026-10-10T03:48:18.285Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).