ScholarGate
Assistente

Confronta i metodi

Esamina i metodi selezionati fianco a fianco; le righe che differiscono sono evidenziate.

TSMixer: Architettura interamente MLP per la previsione di serie temporali×TimeMixer: Decomposable Multiscale Mixing per le Previsioni di Serie Storiche×
CampoApprendimento profondoApprendimento profondo
FamigliaMachine learningMachine learning
Anno di origine20232024
IdeatoreSi-An Chen et al. (Google)Shiyu Wang et al.
TipoAll-MLP multivariate time-series forecasting modelMLP-based multiscale time-series forecasting model
Fonte seminaleChen, 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. link ↗Wang, S., Wu, H., Shi, X., Hu, T., Luo, H., Ma, L., Zhang, J. Y., & Zhou, J. (2024). TimeMixer: Decomposable multiscale mixing for time series forecasting. ICLR. link ↗
AliasAll-MLP Time Series Mixer, Time Series Mixer, TSMixer Forecasting Model, Zaman Serisi KarıştırıcıDecomposable Multiscale Mixing, Multiscale Time-Series Mixer, TimeMixer Model, Çok Ölçekli Zaman Serisi Karıştırıcı
Correlati33
SintesiTSMixer 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.TimeMixer is a decomposition-based, attention-free time-series forecasting architecture introduced by Wang et al. at ICLR 2024. The central idea is to disentangle seasonal and trend components across multiple temporal scales constructed by average pooling, then mix information across those scales using lightweight MLP blocks. By handling coarse (trend-dominant) and fine (seasonal-dominant) resolutions separately and combining their predictions, TimeMixer avoids the quadratic cost of attention while capturing both local and global temporal patterns.
ScholarGateInsieme di dati
  1. v1
  2. 1 Fonti
  3. PUBLISHED
  1. v1
  2. 1 Fonti
  3. PUBLISHED

Vai alla ricerca Scarica le diapositive

ScholarGateConfronta i metodi: TSMixer · TimeMixer. Consultato il 2026-06-17 da https://scholargate.app/it/compare