Our proposed method, called Water-wave Information Transmission Recurrent Acceleration Network (WITRAN), outperforms the state-of-the-art methods by 5.80% and 14.28% on long-range and ultra-long-range time series forecasting tasks respectively, as demonstrated by experiments on four benchmark datasets.
Our paper, titled WITRAN: Water-wave Information Transmission and Recurrent Acceleration Network for Long-range Time Series Forecasting, has been accepted at NeurIPS 2023 as a spotlight! The final version can be found at: [WITRAN].
- Install Python>=3.9, PyTorch 1.10.1.
- Download data. You can obtain all the benchmark datastes from [Autoformer] or [Informer].
- Train the model. Please change the default dataset and parameters in
run.pyand execute it with the following command:
python run.py@inproceedings{jia2023witran,
title={WITRAN: Water-wave Information Transmission and Recurrent Acceleration Network for Long-range Time Series Forecasting},
author={Jia, Yuxin and Lin, Youfang and Hao, Xinyan and Lin, Yan and Guo, Shengnan and Wan, Huaiyu},
booktitle={Thirty-seventh Conference on Neural Information Processing Systems},
year={2023}
}