قارن الطرق

	TSMixer: بنية تعتمد كليًا على طبقات MLP للتنبؤ بالسلاسل الزمنية	TimeMixer: مزج متعدد المقاييس قابل للتحلل للتنبؤ بالسلاسل الزمنية
المجال	التعلم العميق	التعلم العميق
العائلة	Machine learning	Machine learning
سنة النشأة	2023	2024
صاحب الطريقة	Si-An Chen et al. (Google)	Shiyu Wang et al.
النوع	All-MLP multivariate time-series forecasting model	MLP-based multiscale time-series forecasting model
المصدر التأسيسي	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.	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.
الأسماء البديلة	All-MLP Time Series Mixer, Time Series Mixer, TSMixer Forecasting Model, Zaman Serisi Karıştırıcı	Decomposable Multiscale Mixing, Multiscale Time-Series Mixer, TimeMixer Model, Çok Ölçekli Zaman Serisi Karıştırıcı
ذات صلة	3	3
الملخص	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.	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.