发表机构
College of Science, Northwest A&F University（西北农林科技大学理学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对联邦场景下客户端漂移与采样方差问题，提出 FedZOO、FedVRZO 两种联邦零阶优化算法，建立收敛性保证，在黑盒对抗攻击中验证了有效性。
AI 中文摘要
我们研究联邦场景下的约束非光滑非凸随机优化问题，其中客户端仅能获取随机函数评估值。现有联邦零阶方法主要将局部零阶更新与模型平均相结合，但它们难以应对局部投影更新引发的客户端漂移，以及随机零阶估计中的采样方差问题。本文提出一种基于方向聚合的联邦零阶框架，配备两种局部采样方案：具体而言，FedZOO 从跨多个球形查询共享的小批量数据中构建局部方向，从而控制由数据采样和梯度近似产生的随机误差；相比之下，FedVRZO 从独立的样本-方向对中形成局部估计量，因此其采样误差可直接由样本-方向对的数量控制。在这两种算法中，客户端计算沿其投影局部轨迹评估的方向的均值，服务器则使用客户端方向的加权聚合执行单次投影更新。此外，我们分别为 FedZOO 和 FedVRZO 建立了收敛性保证。对黑盒对抗攻击的实验证明了所提方法的有效性。
英文摘要
We study constrained nonsmooth nonconvex stochastic optimization in federated settings, where clients access only stochastic function evaluations. Existing federated zeroth-order methods primarily combine local zeroth-order updates with model averaging. However, they struggle with client drift induced by local projected updates and sampling variance in stochastic zeroth-order estimates. In this paper, we propose a federated zeroth-order framework based on direction aggregation, equipped with two local sampling schemes. Specifically, FedZOO constructs a local direction from a minibatch shared across multiple spherical queries, thereby controlling the stochastic errors arising from data sampling and gradient approximation. In contrast, FedVRZO forms its local estimator from independent sample--direction pairs, so that its sampling error is controlled directly by the number of sample--direction pairs. In both algorithms, clients compute the mean of the directions evaluated along their projected local trajectories, and the server performs a single projected update using the weighted aggregate of the client directions. Furthermore, we establish convergence guarantees for FedZOO and FedVRZO, respectively. Experiments on black-box adversarial attacks demonstrate the effectiveness of the proposed methods.
Comments16 pages
