HiMoE: Heterogeneity-Informed Mixture-of-Experts for Fair Spatial-Temporal Forecasting
This addresses fairness issues in spatial-temporal forecasting for applications requiring reliable outcomes across diverse nodes, representing a novel method for a known bottleneck.
The paper tackles the problem of achieving both accurate and consistent predictive performance across heterogeneous spatial nodes in fair spatial-temporal forecasting by proposing the HiMoE framework, which outperforms the best baseline by at least 9.22% across all evaluation metrics.
Achieving both accurate and consistent predictive performance across spatial nodes is crucial for ensuring the validity and reliability of outcomes in fair spatial-temporal forecasting tasks. However, existing training methods treat heterogeneous nodes with a fully averaged perspective, resulting in inherently biased prediction targets. Balancing accuracy and consistency is particularly challenging due to the multi-objective nature of spatial-temporal forecasting. To address this issue, we propose a novel Heterogeneity-Informed Mixture-of-Experts (HiMoE) framework that delivers both uniform and precise spatial-temporal predictions. From a model architecture perspective, we design the Heterogeneity-Informed Graph Convolutional Network (HiGCN) to address trend heterogeneity, and we introduce the Node-wise Mixture-of-Experts (NMoE) module to handle cardinality heterogeneity across nodes. From an evaluation perspective, we propose STFairBench, a benchmark that handles fairness in spatial-temporal prediction from both training and evaluation stages. Extensive experiments on four real-world datasets demonstrate that HiMoE achieves state-of-the-art performance, outperforming the best baseline by at least 9.22% across all evaluation metrics.