联邦学习中实现干预推断，提升政策评估精度。
Fed-CausalDiff: Decoupled Synchronization for Federated Do-Simulation and Policy Evaluation
- 拆分全局因果与局部混杂得分，实现解耦同步。
- 在4个数据集上显著提升平均处理效应与策略价值估计精度。
- 适合需要低通信开销的跨机构政策评估场景。
联邦学习虽能协同建模分散数据，但传统方法仅拟合历史观测，无法支持干预推断与政策评估，因序列动作会动态改变未来状态。本文提出Fed-CausalDiff，一种用于do-模拟的联邦因果扩散框架。该架构将潜在状态演化分解为全局因果得分函数与局部混杂得分函数，实现解耦同步（DSS）：客户端仅聚合共享因果机制，本地保留站点特异性混杂因素以应对异质性。在四个数据集上的实验表明，Fed-CausalDiff在平均处理效应（ATE）与策略价值估计上均表现更优，同时在通信成本与推断保真度间取得良好权衡。
原文摘要 · Abstract (English)
While federated learning enables collaborative modelling on decentralised data, standard methods merely fit historical observations. This purely observational approach is fundamentally insufficient for interventional inference and policy evaluation, as sequential actions dynamically alter future states. We propose \textbf{Fed-CausalDiff}, a federated causal diffusion framework for do-simulation. The architecture decomposes the evolution of the latent state into a global causal score function and a local confounding score function. This design enables \emph{decoupled synchronisation} (DSS), where clients aggregate only the shared causal mechanism while retaining site-specific confounders locally to handle heterogeneity. Experiments on four datasets demonstrate that Fed-CausalDiff achieves better ATE and policy-value estimation accuracy, offering a favorable trade-off between communication cost and inference fidelity.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。