Machine learningTime-series forecasting
iTransformer：用于多元时间序列预测的倒置Transformer
iTransformer是由Liu等人于ICLR 2024上提出的一种用于多元时间序列预测的深度学习架构。其核心思想是颠覆传统的Transformer分词策略：iTransformer不将每个时间步视为一个token，而是将每个变量（传感器通道或特征序列）视为一个单独的token，其嵌入编码了完整的观测回溯窗口。然后，自注意力机制应用于变量之间，以捕获序列间的依赖关系，而每个token内的前馈网络则学习时间模式。
阅读完整方法
仅限会员
登录使用免费账户登录即可阅读本节。
Method map
The neighbourhood of related methods — select a node to explore.
来源
- Liu, Y., Hu, T., Zhang, H., Wu, H., Wang, S., Ma, L., & Long, M. (2024). iTransformer: Inverted transformers are effective for time series forecasting. ICLR. link ↗
如何引用本页
ScholarGate. (2026, June 2). iTransformer (Inverted Transformer for Forecasting). ScholarGate. https://scholargate.app/zh/deep-learning/itransformer
Which method?
Set this method beside its closest kin and read them side by side — the library lays the books on the table; the choice is yours.
Compare side by side →