论文

ReMoE:可解释混合专家框架用于跨灾害人群流动恢复预测

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

精选理由

用 mixture-of-experts 学跨灾害共有的恢复模式,18 个灾害数据上验证。可解释这点很实用——你能看到每个区域被分给哪类专家。做灾害响应、城市数据或时空预测的工程师值得跑一遍代码。

论文提出 ReMoE,一个可解释的 mixture-of-experts 框架,用于预测灾害后人群流动的恢复过程。该框架的核心是一个 feature-guided router,自适应地组合多个专门学习不同恢复类型的专家,生成针对特定区域的恢复模式,并通过未来灾害条件预测信息进行残差修正。

正文

评估覆盖 18 个灾害事件的大规模流动数据集,涉及飓风、洪水、冬季风暴和野火。结果显示 ReMoE 在分布内表现有竞争力,在未见过的灾害事件和灾害类型上取得最强性能,且分布偏移越大提升越明显。

区别于以往针对单一事件建模的黑盒方法,ReMoE 的专家划分具有可解释的学习结果,恢复知识可跨灾害迁移,能在不确定性下保持稳健表现,可用于灾后预判性评估和资源分配。

原文 · arXiv cs.LG

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

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.

产业坐标与信源核验

本条情报归属 AITOP「论文」核心追踪栏目。涉及核心实体与技术议题包括:ReMoE、Mixture-of-Experts、时空预测、灾害响应,读者可通过下方标签进入对应专题页查阅历史脉络与关联对比。

信源核验说明:本文原始线索采编自「arXiv cs.LG」(@Yilin Lyu, Kang Liu, Wei Wang, Qifan Zhang, Ling Yin, Yushi Bai),经 AITOP 多维评分管线结构化提炼。