Heterogeneity-Informed Meta-Parameter Learning for Spatiotemporal Time Series Forecasting
This work addresses spatiotemporal forecasting problems for applications like traffic or weather prediction, but it appears incremental as it builds on existing meta-learning and embedding techniques.
The paper tackles the challenge of capturing spatiotemporal heterogeneity in time series forecasting by proposing a Heterogeneity-Informed Meta-Parameter Learning scheme, which achieves state-of-the-art performance on five benchmarks with improved interpretability.
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.