Metraipolis › archive › dc_23e43bb04858
LagNet: Lag-Structured Operators for Long-Term Time Series Forecasting — arXiv:2610.23735
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.23735
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-19T09:10:27.008Z
- SHA-256 of the archived text
fa53ee467d5d5d904cc809034d5138793a1228295eece976c05ea18e1be9b465- Stable URL of this page
https://metraipolis.com/documento/dc_23e43bb04858
How to cite
LagNet: Lag-Structured Operators for Long-Term Time Series Forecasting — arXiv:2610.23735 — arXiv (en), https://arxiv.org/abs/2610.23735 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_23e43bb04858 (Metraipolis archive dc_23e43bb04858, sha256 fa53ee467d5d5d904cc809034d5138793a1228295eece976c05ea18e1be9b465)
BibTeX
@misc{dc23e43bb04858,
title = {LagNet: Lag-Structured Operators for Long-Term Time Series Forecasting — arXiv:2610.23735},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.23735},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_23e43bb04858; sha256 fa53ee467d5d5d904cc809034d5138793a1228295eece976c05ea18e1be9b465},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.23735. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Long-term time series forecasting maps a finite history window to an extended future horizon. Many existing models focus primarily on the design of input representations and model structures, while paying less attention to the structural form of the history-to-future forecasting operator itself. We examine this operator directly and find that, across model families, sensitivity matrices exhibit recurring diagonal patterns whose diagonals correspond to relative temporal lags. Motivated by this observation, we develop a framework for direct multi-step forecasting based on lag-structured temporal operators. Its basic building block, LagLayer, constructs a Toeplitz matrix by tying history-to-future connections with the same relative lag and incorporates horizon-wise and history-wise scaling coefficients to relax exact sharing. This design enables LagLayer to serve as an effective replacement for temporal linear layers in existing forecasting models. Stacking LagLayer modules with activations and multiple feature maps yields LagNet. Experiments on long-term time series forecasting benchmarks show that our method is competitive with strong baselines, while sensitivity matrix diagnostics and limited-data experiments demonstrate that LagLayer provides an effective and sample-efficient alternative to a linear layer. 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/2610.23735 — 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 2026-10-19T09:06:29.570Z dalla pagina https://arxiv.org/list/cs.LG/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).