Metraipolis › archive › dc_31b2ef52ae68
Learning From Static Context in Federated Time Series Forecasting — arXiv:2611.06064
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.06064
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-05T18:22:08.671Z
- SHA-256 of the archived text
8b560932386f91cce205a8a0b1c9a9af7134003718012b5fbdaa00e4f98e4584- Stable URL of this page
https://metraipolis.com/documento/dc_31b2ef52ae68
How to cite
Learning From Static Context in Federated Time Series Forecasting — arXiv:2611.06064 — arXiv (en), https://arxiv.org/abs/2611.06064 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_31b2ef52ae68 (Metraipolis archive dc_31b2ef52ae68, sha256 8b560932386f91cce205a8a0b1c9a9af7134003718012b5fbdaa00e4f98e4584)
BibTeX
@misc{dc31b2ef52ae68,
title = {Learning From Static Context in Federated Time Series Forecasting — arXiv:2611.06064},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.06064},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_31b2ef52ae68; sha256 8b560932386f91cce205a8a0b1c9a9af7134003718012b5fbdaa00e4f98e4584},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.06064. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Time series forecasting is shifting from locally trained, per-series models to global models trained across many heterogeneous series, which requires learning what sets the series apart. When the raw data cannot be centralized, such a global model is trained with federated learning (FL), one client per data owner. State-of-the-art forecasters such as the Temporal Fusion Transformer (TFT) tell series apart through a static input that describes each series and does not change over time. In the special case of a federated setting where each client holds a single series, this static input is constant within each client and varies only across clients. We ask whether a federated TFT can learn from it. Per-batch FedAdam, which aggregates gradients after every batch, uses the static input almost as well as central training, whereas communication-efficient schemes with local epochs (FedAvg, FedAdam+Prox) extract far less from it and gain nothing significant on either application dataset. At the same local computation, an Adam-compatible SCAFFOLD keeps the static pathway responsive on wind power and gains significantly there. On building energy its error still drops, but the static pathway falls silent, so drift and communication frequency cannot be fully separated as the cause. Only per-batch FedAdam matches or beats the per-client local models across all datasets. We confirm the effect on building-energy meter data, a wind-power dataset, and a controlled synthetic dataset. The loss is not specific to the TFT: on the synthetic and the wind data two further architectures with different static-conditioning mechanisms show it as well. 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/2611.06064 — 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 2026-11-05T18:15:49.011Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).