arXiv：cs.LG· Ayush Mohanty, Nazal Mohamed, Nagi Gebraeel·· 3 小时前
联邦 Granger 因果学习中的不确定性量化
Uncertainty Quantification in Federated Granger Causality Learning
AI 导读
该论文刻画了联邦 Granger 因果学习中的不确定性传播，推导出客户端-服务器迭代的闭式方差递推与稳态方差，并证明初始模型参数不确定性的传播贡献渐近消失。基于边特异方差，方法可用于统计上区分真实的跨客户端依赖与虚假边。合成实验显示其跨客户端边恢复优于对比基线，在真实工业数据集上取得高根因识别准确率并给出可解释的依赖结构。
正文
Abstract:Granger causality identifies predictive dependencies in multivariate time series. In distributed settings where parties cannot share data, federated causal learning enables joint analysis. Most federated causal methods assume that clients observe the same features and infer causal relationships as point estimates, with little formal uncertainty quantification. These assumptions do not hold in many industrial systems, where clients observe different features, and the objective is to estimate cross-client dependencies (edges). These dependencies must be estimated indirectly through repeated client-server iterations. Uncertainty from client data and model parameters propagates through this process, making point estimates alone insufficient for assessing cross-client edges. This paper characterizes this uncertainty propagation and uses edge-specific variances to distinguish genuine cross-client dependencies from spurious estimated edges. We consider aleatoric uncertainty from client data variability and epistemic uncertainty from model parameters. We derive closed-form variance recursions and steady-state variances for the client-server iterations. We prove that the propagated contribution of the initial model-parameter uncertainty vanishes asymptotically. These variances enable statistically principled selection of cross-client edges. Synthetic experiments show that our approach improves cross-client edge recovery over competing baselines. On real-world industrial datasets, it achieves high root-cause identification accuracy while yielding interpretable dependency structures.
| Comments: | Manuscript under review |
| Subjects: | Machine Learning (cs.LG); Machine Learning (stat.ML) |
| ACM classes: | I.2 |
| Cite as: | arXiv:2602.13004 [cs.LG] |
| (or arXiv:2602.13004v3 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2602.13004 arXiv-issued DOI via DataCite |
Submission history
From: Ayush Mohanty [view email]
[v1]
Fri, 13 Feb 2026 15:12:18 UTC (31,424 KB)
[v2]
Mon, 11 May 2026 19:11:15 UTC (14,089 KB)
[v3]
Thu, 8 Oct 2026 15:49:51 UTC (27,495 KB)
来源：arXiv：cs.LG · arxiv.org