Metraipolis › archive › dc_d75bcb7dbda0

Learning to Difference: Adaptive Reversible Differencing (AdaRDiff) for Time Series Forecasting — arXiv:2608.28134

Source: arXiv (en)
Original: https://arxiv.org/abs/2608.28134
License (as archived): CC0-1.0
Archived: 2026-10-17T11:06:32.327Z
SHA-256 of the archived text: bebb9a931d171dc7990f4a3b17f79bd74d56cd0d1895e9214257a1082293bd96
Stable URL of this page: https://metraipolis.com/documento/dc_d75bcb7dbda0

How to cite

Learning to Difference: Adaptive Reversible Differencing (AdaRDiff) for Time Series Forecasting — arXiv:2608.28134 — arXiv (en), https://arxiv.org/abs/2608.28134 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d75bcb7dbda0 (Metraipolis archive dc_d75bcb7dbda0, sha256 bebb9a931d171dc7990f4a3b17f79bd74d56cd0d1895e9214257a1082293bd96)

BibTeX

@misc{dcd75bcb7dbda0,
  title = {Learning to Difference: Adaptive Reversible Differencing (AdaRDiff) for Time Series Forecasting — arXiv:2608.28134},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.28134},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d75bcb7dbda0; sha256 bebb9a931d171dc7990f4a3b17f79bd74d56cd0d1895e9214257a1082293bd96},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.28134. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Reliable long-horizon time series forecasting is an important yet difficult problem. Trends and seasonality introduce complex temporal structure that challenges learning-based forecasting models. Differencing, which subtracts nearby past values to remove such structure, is the classical remedy, but its reliance on hand-picked orders and periods has kept it largely absent from recent deep architectures. We propose \textbf{\underline{Ada}}ptive \textbf{\underline{R}}eversible \textbf{\underline{Diff}}erencing \textbf{(AdaRDiff)}, a generalized differencing approach that uses learnable weights to simplify the series through weighted differencing with previous time instants. This yields stabilized residuals on which forecasting is performed, after which the removed components are restored autoregressively to reconstruct the forecast, capturing trend and seasonality jointly through a single operator. This reconstruction admits a closed-form convolutional expression, which parallelizes on GPU and yields up to $33.7\times$ speedup over the naive recurrence. We furthermore rely on a two-phase training schedule that separates temporal structure discovery from reconstruction learning, as suggested by a theoretical analysis of the gradient when using a linear forecasting model. AdaRDiff attains state-of-the-art forecast accuracy across eight benchmarks spanning electricity, weather, traffic, and energy, at negligible parameter cost. Furthermore, it is designed as a plug-and-play module: integrating AdaRDiff improves eight diverse backbones, from linear models to Transformers, in the large majority of cases, by up to $25.9\%$ with a linear backbone and $18.3\%$ with iTransformer.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-31 in cs.LG).
Categorie arXiv: Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2608.28134 — 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 2026-10-17T10:50:25.796Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-08-31?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).