WITRAN: Water-wave Information Transmission and Recurrent Acceleration Network for Long-range Time Series Forecasting

Capturing semantic information is crucial for accurate long-range time series forecasting, which involves modeling global and local correlations, as well as discovering long-and short-term repetitive patterns. Previous works have partially addressed these issues separately, but have not been able to address all of them simultaneously. Meanwhile, their time and memory complexities are still not sufficiently low for long-range forecasting. To address the challenge of capturing different types of semantic information, we propose a novel Water-wave Information Transmission (WIT) framework. This framework captures both long-and short-term repetitive patterns through bi-granular information transmission. It also models global and local correlations by recursively fusing and selecting information using Horizontal Vertical Gated Selective Unit (HVGSU). In addition, to improve the computing efficiency, we propose a generic Recurrent Acceleration Network (RAN) which reduces the time complexity to O( √ L) while maintaining the memory complexity at O(L). Our proposed method, called Water-wave Information Transmission and 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. The code is available at: https://github.com/Water2sea/WITRAN . * Corresponding author 37th Conference on Neural Information Processing Systems (NeurIPS 2023). et al., 2018 et al., , Liu et al., 2021]] , such as hourly or daily cycles, which can be identified as periodic semantic information. Furthermore, it is crucial to model both aspects of semantic information simultaneously and ensure that the modeling approach remains computationally efficient, avoiding excessive complexity. Unfortunately, although the existing state-of-the-art methods have shown great performance, they encounter difficulties in addressing the aforementioned challenges simultaneously. Specifically, CNN-based methods [