Heterogeneity-Informed Meta-Parameter Learning for Spatiotemporal Time Series Forecasting
Dong, Zheng, Jiang, Renhe, Gao, Haotian, Liu, Hangchen, Deng, Jinliang, Wen, Qingsong, Song, Xuan
–arXiv.org Artificial Intelligence
Spatiotemporal time series forecasting plays a key role in a wide range of real-world applications. While significant progress has been made in this area, fully capturing and leveraging spatiotemporal heterogeneity remains a fundamental challenge. Therefore, we propose a novel Heterogeneity-Informed Meta-Parameter Learning scheme. Specifically, our approach implicitly captures spatiotemporal heterogeneity through learning spatial and temporal embeddings, which can be viewed as a clustering process. Then, a novel spatiotemporal meta-parameter learning paradigm is proposed to learn spatiotemporal-specific parameters from meta-parameter pools, which is informed by the captured heterogeneity. Based on these ideas, we develop a Heterogeneity-Informed Spatiotemporal Meta-Network (HimNet) for spatiotemporal time series forecasting. Extensive experiments on five widely-used benchmarks demonstrate our method achieves state-of-the-art performance while exhibiting superior interpretability. Our code is available at https://github.com/XDZhelheim/HimNet.
arXiv.org Artificial Intelligence
May-17-2024
- Country:
- North America
- Europe > Spain
- Catalonia > Barcelona Province > Barcelona (0.05)
- Asia
- China > Hong Kong (0.04)
- Japan > Honshū
- Tōhoku > Fukushima Prefecture
- Fukushima (0.04)
- Kantō > Tokyo Metropolis Prefecture
- Tokyo (0.04)
- Tōhoku > Fukushima Prefecture
- Genre:
- Research Report (0.64)
- Industry:
- Transportation > Ground > Road (0.46)
- Technology: