Vertaile menetelmiä

	ETSformer: Eksponentiaalisen tasoituksen muuntajat aikasarjaennustamiseen	Autoformer: Dekompositio-muuntaja pitkän aikavälin aikasarjaennustamiseen	ETS: Virhe-, trendi- ja kausitasoitus
Tieteenala	Syväoppiminen	Syväoppiminen	Ekonometria
Menetelmäperhe	Machine learning	Machine learning	Regression model
Syntyvuosi	2022	2021	2008
Kehittäjä	Gerald Woo et al.	Haixu Wu et al. (Tsinghua)	Hyndman, Koehler, Ord & Snyder (state space framework)
Tyyppi	Hybrid decomposition-based Transformer architecture	Decomposition-based deep forecasting model	Exponential smoothing state space model
Alkuperäislähde	Woo, G., Liu, C., Sahoo, D., Kumar, A., & Hoi, S. (2022). ETSformer: Exponential smoothing transformers for time-series forecasting. arXiv preprint.	Wu, H., Xu, J., Wang, J., & Long, M. (2021). Autoformer: Decomposition transformers with auto-correlation for long-term series forecasting. NeurIPS, 34.	Hyndman, R. J., Koehler, A. B., Ord, J. K. & Snyder, R. D. (2008). Forecasting with Exponential Smoothing: The State Space Approach. Springer.
Rinnakkaisnimet	Exponential Smoothing Transformer, ETS Transformer, ETSformer forecasting model, Üstel Düzleştirme Transformatörü	Auto-Correlation Transformer, Decomposition Transformer, Series Decomposition Forecaster, Oto-Korelasyon Ayrışım Transformer	exponential smoothing state space model, innovations state space model, Holt-Winters family, ETS — Hata/Trend/Mevsimsellik Üstel Düzleştirme
Liittyvät	2	4	5
Tiivistelmä	ETSformer is a deep learning architecture for time-series forecasting introduced by Woo et al. in 2022. It integrates classical exponential smoothing principles directly into the Transformer framework by replacing standard self-attention with an exponential smoothing attention mechanism. The model decomposes a time series into level, growth (trend), and seasonal components, allowing it to leverage both the long-range dependency modeling of Transformers and the interpretable structure of statistical ETS models.	Autoformer is a deep learning architecture for long-term time-series forecasting, introduced by Wu et al. from Tsinghua University at NeurIPS 2021. It replaces the standard self-attention mechanism with an Auto-Correlation mechanism that exploits periodic dependencies in the frequency domain, and embeds a progressive series decomposition block throughout the encoder and decoder to separately model trend and seasonal components.	ETS is a comprehensive exponential smoothing framework that automatically selects additive or multiplicative combinations of the error (E), trend (T) and seasonal (S) components of a time series. Formalised as an innovations state space model by Hyndman, Koehler, Ord and Snyder in 2008, it unifies and generalises the Holt-Winters family of forecasting methods.