[论文解读] Broadband Analog Aggregation for Low-Latency Federated Edge Learning (Extended Version)

Guangxu Zhu, Yong Wang+1 more|arXiv (Cornell University)|Dec 30, 2018

Advanced MIMO Systems Optimization参考文献 34被引用 38

一句话总结

本文提出宽带模拟聚合（BAA），一种用于联邦边缘学习的低延迟多址方案，通过在多址信道上利用波形叠加实现模型更新的空中聚合。BAA通过支持同时传输和模拟聚合，几乎线性地减少了设备数量增加带来的通信延迟，显著优于传统OFDMA，同时在单小区随机网络中量化了可靠性、截断与学习效率之间的理论权衡。

ABSTRACT

The popularity of mobile devices results in the availability of enormous data and computational resources at the network edge. To leverage the data and resources, a new machine learning paradigm, called edge learning, has emerged where learning algorithms are deployed at the edge for providing fast and intelligent services to mobile users. While computing speeds are advancing rapidly, the communication latency is becoming the bottleneck of fast edge learning. To address this issue, this work is focused on designing a low latency multi-access scheme for edge learning. We consider a popular framework, federated edge learning (FEEL), where edge-server and on-device learning are synchronized to train a model without violating user-data privacy. It is proposed that model updates simultaneously transmitted by devices over broadband channels should be analog aggregated "over-the-air" by exploiting the superposition property of a multi-access channel. Thereby, "interference" is harnessed to provide fast implementation of the model aggregation. This results in dramatical latency reduction compared with the traditional orthogonal access (i.e., OFDMA). In this work, the performance of FEEL is characterized targeting a single-cell random network. First, due to power alignment between devices as required for aggregation, a fundamental tradeoff is shown to exist between the update-reliability and the expected update-truncation ratio. This motivates the design of an opportunistic scheduling scheme for FEEL that selects devices within a distance threshold. This scheme is shown using real datasets to yield satisfactory learning performance in the presence of high mobility. Second, both the multi-access latency of the proposed analog aggregation and the OFDMA scheme are analyzed. Their ratio, which quantifies the latency reduction of the former, is proved to scale almost linearly with device population.

研究动机与目标

• 解决因在无线信道上传输高维模型更新而引起的联邦边缘学习（FEEL）通信瓶颈问题。
• 设计一种低延迟多址方案，实现在边缘服务器对模型更新的快速聚合。
• 量化宽带无线环境中通信与学习指标之间的权衡。
• 与传统OFDMA相比，评估BAA在延迟降低、可靠性与学习效率方面的性能表现。
• 探索提升对对抗性攻击鲁棒性的扩展方法，并与波 beamforming 结合以改善小区边缘性能。

提出的方法

• 提出宽带模拟聚合（BAA），多个边缘设备在共享宽带信道上同时传输其模型更新。
• 利用多址信道的波形叠加特性，实现在边缘服务器对更新的模拟聚合，避免使用正交接入（如OFDMA）。
• 采用截断信道反转功率控制，以平衡更新可靠性（信噪比，SNR）与更新截断比例。
• 将设备位置建模为在圆形小区内独立同分布的均匀分布，基于设备到边缘服务器的距离进行调度以提高可靠性。
• 利用条件期望与顺序统计方法，推导出全包含与小区内部调度方案下的期望接收信噪比。
• 使用广义Marcum Q函数与指数积分函数，对信噪比表达式中的衰落与路径损耗效应进行建模。

实验结果

研究问题

• RQ1 与传统OFDMA相比，BAA在联邦边缘学习中如何降低通信延迟？
• RQ2 BAA中更新可靠性（信噪比）与更新截断比例、学习效率之间存在何种权衡？
• RQ3 小区内部调度如何影响接收信噪比与学习中所用数据比例之间的权衡？
• RQ4 延迟降低的缩放行为如何随边缘设备数量变化？
• RQ5 如何扩展BAA以增强对对抗性攻击的鲁棒性，并提升小区边缘性能？

主要发现

• 与OFDMA相比，BAA的延迟降低比率几乎随边缘设备数量线性增长，表明在大规模部署中具有显著加速效果。
• 由于采用截断信道反转功率控制，BAA在更新可靠性（以接收信噪比衡量）与期望更新截断比例之间引入了权衡。
• 小区内部调度导致接收信噪比与学习中所用数据比例之间存在权衡，后者取决于小区半径与设备分布。
• 全包含BAA下的期望接收信噪比为 $\mathbb{E}(\rho_0) = \frac{2K}{2K - \alpha} \frac{P_0}{MR^\alpha \text{Ei}(g_{\text{th}})}$，当 $2K - \alpha - 1 \geq 0$ 时收敛。
• 对于小区内部调度，期望信噪比表示为 $\frac{2k}{2k - \alpha}$ 在 $k \geq 2$ 上的加权平均 $c(R_{\text{in}})$，当 $\alpha = 3$ 且 $K$ 较大时，有 $1 \leq c(R_{\text{in}}) \leq 4$。
• 使用神经网络与真实数据集的实验验证了理论结果，证实了BAA在实际场景中的性能优势。