Comparer des méthodes

	FEDformer : Transformeur décomposé à fréquence améliorée	TSMixer : une architecture tout MLP pour la prévision de séries temporelles
Domaine	Apprentissage profond	Apprentissage profond
Famille	Machine learning	Machine learning
Année d'origine	2022	2023
Auteur d'origine	Tian Zhou et al.	Si-An Chen et al. (Google)
Type	Frequency-domain decomposed Transformer for time-series forecasting	All-MLP multivariate time-series forecasting model
Source fondatrice	Zhou, T., Ma, Z., Wen, Q., Wang, X., Sun, L., & Jin, R. (2022). FEDformer: Frequency enhanced decomposed transformer for long-term series forecasting. ICML.	Chen, S.-A., Li, C.-L., Yoder, N., Arik, S. O., & Pfister, T. (2023). TSMixer: An all-MLP architecture for time series forecasting. Transactions on Machine Learning Research.
Alias	Frequency Enhanced Decomposed Transformer, FED-Transformer, Frequency Domain Transformer, Frekans Tabanlı Ayrıştırılmış Dönüştürücü	All-MLP Time Series Mixer, Time Series Mixer, TSMixer Forecasting Model, Zaman Serisi Karıştırıcı
Apparentées	3	3
Résumé	FEDformer is a Transformer-based architecture for long-term multivariate time-series forecasting, introduced by Zhou et al. at ICML 2022. Its core innovation is the combination of seasonal-trend decomposition with frequency-domain attention: instead of computing full token-to-token attention in the time domain, FEDformer projects queries, keys, and values into the frequency domain via Fourier or wavelet transforms and operates on a randomly selected subset of frequency components, achieving linear complexity while preserving global temporal structure.	TSMixer is a multivariate time-series forecasting model introduced by Si-An Chen and colleagues at Google in 2023. It challenges the prevailing dominance of Transformer-based architectures by demonstrating that a simple stack of interleaved MLP layers — alternating between mixing along the time axis and mixing across feature channels — achieves strong forecasting accuracy while remaining computationally efficient and easy to interpret architecturally.