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.
@article{arxiv.2405.10800,
title = {Heterogeneity-Informed Meta-Parameter Learning for Spatiotemporal Time Series Forecasting},
author = {Zheng Dong and Renhe Jiang and Haotian Gao and Hangchen Liu and Jinliang Deng and Qingsong Wen and Xuan Song},
journal= {arXiv preprint arXiv:2405.10800},
year = {2024}
}