MSHyper: Multi-Scale Hypergraph Transformer for Long-Range Time Series Forecasting
This addresses a bottleneck in time series forecasting for applications requiring long-range predictions, though it appears incremental as it builds on existing hypergraph and transformer methods.
The paper tackled the problem of modeling high-order interactions between temporal patterns at different scales for long-range time series forecasting by proposing the MSHyper framework, which achieved state-of-the-art performance on five real-world datasets.
Demystifying interactions between temporal patterns of different scales is fundamental to precise long-range time series forecasting. However, previous works lack the ability to model high-order interactions. To promote more comprehensive pattern interaction modeling for long-range time series forecasting, we propose a Multi-Scale Hypergraph Transformer (MSHyper) framework. Specifically, a multi-scale hypergraph is introduced to provide foundations for modeling high-order pattern interactions. Then by treating hyperedges as nodes, we also build a hyperedge graph to enhance hypergraph modeling. In addition, a tri-stage message passing mechanism is introduced to aggregate pattern information and learn the interaction strength between temporal patterns of different scales. Extensive experiments on five real-world datasets demonstrate that MSHyper achieves state-of-the-art (SOTA) performance across various settings.