Porównaj metody

	N-BEATS	Temporal Fusion Transformer
Dziedzina	Uczenie głębokie	Uczenie głębokie
Rodzina	Machine learning	Machine learning
Rok powstania	2020	2021
Twórca	Oreshkin, B.N. et al.	Lim, B., Arık, S. Ö., Loeff, N. & Pfister, T.
Typ	Deep neural forecasting architecture (interpretable basis expansion)	Attention-based deep learning forecasting architecture
Źródło pierwotne	Oreshkin, B.N. et al. (2020). N-BEATS: Neural Basis Expansion Analysis for Interpretable Time Series Forecasting. ICLR.	Lim, B., Arık, S. Ö., Loeff, N. & Pfister, T. (2021). Temporal Fusion Transformers for Interpretable Multi-Horizon Time Series Forecasting. International Journal of Forecasting, 37(4), 1748–1764.
Inne nazwy	N-BEATS — Nöral Zaman Serisi Tahmini, Neural Basis Expansion Analysis, neural basis expansion	Temporal Fusion Transformer (TFT), TFT, interpretable multi-horizon forecasting transformer
Pokrewne	5	6
Podsumowanie	N-BEATS is a deep learning architecture for time series forecasting, introduced by Oreshkin and colleagues in 2020, built from interpretable trend and seasonality stacks. It was the first purely neural forecasting model to reach state-of-the-art performance on the M4 competition without relying on any classical statistical components.	The Temporal Fusion Transformer (TFT), introduced by Lim, Arık, Loeff and Pfister in 2021, is an interpretable deep learning architecture for multi-horizon time series forecasting. It combines variable selection, gating, multi-horizon attention and quantile outputs, processing static, past and known-future inputs together to produce multi-step forecasts.