跳到正文
arXiv cs.CY· Yilin Lyu, Kang Liu, Wei Wang, Qifan Zhang, Ling Yin, Yushi Bai·· 9 小时前AI 评分31

ReMoE:面向跨灾害人群移动恢复预测的可解释混合专家框架

ReMoE: An Interpretable Mixture-of-Experts Framework for Human Mobility Recovery Forecasting Across Disasters

AI 导读

ReMoE 是一个可解释的混合专家框架,用于跨灾害的人群移动恢复预测,通过特征引导路由器自适应组合学习不同恢复类型的专家,并用未来灾害条件预测做残差修正。在覆盖飓风、洪水、冬季风暴和野火共 18 起灾害事件的大规模移动数据集上,ReMoE 分布内表现具竞争力,在未见事件和灾害类型上取得最强性能,分布偏移越大增益越明显。该工作已被 KDD 2027 接收。

正文 · 原文

View PDF HTML (experimental)

Abstract:Human mobility often declines sharply following disasters and subsequently exhibits complex recovery dynamics. Effective forecasting of this recovery is critical for timely assessment and resource allocation. However, existing approaches are often developed for individual events and operate as black boxes, limiting their transferability and interpretability beyond the events they were trained on. Recent studies have identified recurring post-disaster mobility recovery types across disasters, suggesting that recovery knowledge may be transferable. Motivated by this, we propose ReMoE, an interpretable mixture-of-experts framework for cross-disaster mobility recovery forecasting. ReMoE uses a feature-guided router to adaptively combine specialized experts that learn these heterogeneous yet reusable recovery types, producing a region-specific recovery pattern. It further refines this pattern through residual correction informed by forecasts of future hazard conditions. We evaluate ReMoE on a large-scale mobility dataset covering 18 disaster events spanning hurricanes, floods, winter storms, and wildfires. ReMoE is competitive in-distribution and achieves the strongest performance on unseen events and hazard types, with larger gains under greater distribution shifts. Further analyses show that ReMoE yields interpretable learning outcomes and maintains robust performance under uncertainty. Broadly, ReMoE offers a transferable and traceable forecaster that can support anticipatory post-disaster assessment.
Comments: Accepted by KDD 2027
Subjects: Machine Learning (cs.LG); Computers and Society (cs.CY)
Cite as: arXiv:2611.22133 [cs.LG]
  (or arXiv:2611.22133v1 [cs.LG] for this version)
  https://doi.org/10.48550/arXiv.2611.22133

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Yilin Lyu [view email]
[v1] Tue, 17 Nov 2026 03:18:34 UTC (11,843 KB)

来源:arXiv cs.CY · arxiv.org