Abstract
Non-IID data and partial participation induce client drift and inconsistent local optima in federated learning, causing unstable convergence and accuracy loss. We present FedSSG, a stochastic sampling-guided, history-aware drift alignment method. FedSSG maintains a per-client drift memory that accumulates local model differences as a lightweight sketch of historical gradients; crucially, it gates both the memory update and the local alignment term by a smooth function of the observed/expected participation ratio (a phase-by-expectation signal derived from the server sampler). This statistically grounded gate stays weak and smooth when sampling noise dominates early, then strengthens once participation statistics stabilize, contracting the local-global gap without extra communication. Across CIFAR-10/100 with 100/500 clients and 2-15 percent participation, FedSSG consistently outperforms strong drift-aware baselines and accelerates convergence; on our benchmarks it improves test accuracy by up to a few points (e.g., about +0.9 on CIFAR-10 and about +2.7 on CIFAR-100 on average over the top-2 baseline) and yields about 4.5x faster target-accuracy convergence on average. The method adds only O(d) client memory and a constant-time gate, and degrades gracefully to a mild regularizer under near-IID or uniform sampling. FedSSG shows that sampling statistics can be turned into a principled, history-aware phase control to stabilize and speed up federated training.
Abstract (translated)
非独立同分布（Non-IID）数据和部分参与会导致联邦学习中的客户端漂移以及局部最优解不一致，从而引发不稳定收敛和准确率下降。我们提出了FedSSG，这是一种基于随机采样指导的、历史感知型漂移对齐方法。FedSSG为每个客户端维护一个漂移记忆，该记忆累积了本地模型差异的历史梯度轻量级草图；关键在于，它通过平滑函数来控制记忆更新和局部对齐项，该函数依赖于观察到/预期的参与比率（由服务器采样器导出的一个期望建模信号）。当抽样噪声在早期占主导时，这个基于统计的方法会保持弱且平滑的状态；一旦参与统计稳定下来，它就会增强力度，缩小局部与全局之间的差距，而无需额外通信。在CIFAR-10/100数据集上使用100/500个客户端和2%-15%的参与度时，FedSSG始终优于强大的漂移感知基准方法，并且加速了收敛过程；在我们的基准测试中，它平均将测试准确率提高了几个百分点（例如，在CIFAR-10上约为+0.9，在CIFAR-100上约为+2.7），并且平均实现了目标准确率4.5倍的更快收敛。该方法仅增加了O(d)大小的客户端内存和一个常数时间的门控机制，并在接近独立同分布（IID）或均匀采样时平稳退化为轻量级正则器。FedSSG表明，抽样统计可以被转化为一种原则性的、历史感知型阶段控制手段，以稳定并加速联邦训练过程。
URL
https://arxiv.org/abs/2509.13895
