Metraipolis › archive › dc_f31188fbca37

TinyCast: Probabilistic Zero-Shot Forecasting with Computed Periodicity — arXiv:2608.15767

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.15767
License (as archived)
CC0-1.0
Archived
2026-10-18T18:04:23.816Z
SHA-256 of the archived text
f945c66f28fdca419fde3464c7a0ea3f0cd429687517ad2ce086e8ed0dbcb272
Stable URL of this page
https://metraipolis.com/documento/dc_f31188fbca37

How to cite

TinyCast: Probabilistic Zero-Shot Forecasting with Computed Periodicity — arXiv:2608.15767 — arXiv (en), https://arxiv.org/abs/2608.15767 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_f31188fbca37 (Metraipolis archive dc_f31188fbca37, sha256 f945c66f28fdca419fde3464c7a0ea3f0cd429687517ad2ce086e8ed0dbcb272)
BibTeX
@misc{dcf31188fbca37,
  title = {TinyCast: Probabilistic Zero-Shot Forecasting with Computed Periodicity — arXiv:2608.15767},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.15767},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_f31188fbca37; sha256 f945c66f28fdca419fde3464c7a0ea3f0cd429687517ad2ce086e8ed0dbcb272},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.15767. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

We introduce TinyCast, an attention-free zero-shot forecaster that emits a predictive distribution from 146,505 parameters, on the premise that at this size the periodic structure of a context is worth computing rather than learning. A zero-parameter spectral detector supplies the dominant periods, the context is folded on their phase, and a dilated convolutional encoder and a block-autoregressive quantile decoder model the rest. It is smaller than every zero-shot entry on the GIFT-Eval board whose parameter count can be established. On probabilistic accuracy it defines the size-accuracy frontier. Among zero-shot entries declaring no test-data leakage it is the only one below 1.4M parameters that emits a predictive distribution, and every entry scoring better carries at least that budget. On Chronos-ZS and fev-bench every neural model ahead of it carries at least 28 times its parameters. Because the mixing path is convolutions and matrix multiplications only, it exports to static INT8 and forecasts end to end on an embedded device without per-signal fitting.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-18 in cs.AI).
Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.15767 — 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 2026-10-18T16:29:14.048Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-18?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).