FDNet: Focal Decomposed Network for Efficient, Robust and Practical Time Series Forecasting
FDNet: Focal Decomposed Network for Efficient, Robust and Practical Time Series Forecasting
Li Shen,Yuning Wei,Yangzhu Wang,Huaxin Qiu
2023 · DOI: 10.1016/j.knosys.2023.110666
Knowledge-Based Systems · 11 Citations
TLDR
By abandoning global coarse-grained feature maps, FDNet overcomes distribution shift problem caused by changing dynamics of time series which is common in real-world applications and reduces prediction MSE by 38.4% on average compared with other thirteen SOTA baselines.