EdgeC3: Online Management for Edge-Cloud Collaborative Continuous Learning
EdgeC3: Online Management for Edge-Cloud Collaborative Continuous Learning
复制标题
DOI：
10.1109/secon58729.2023.10287414
复制
发表时间：
2023-09
期刊：
影响因子：
--
通讯作者：
Shaohui Lin;Xiaoxi Zhang;Yupeng Li;Carlee Joe-Wong;Jingpu Duan;Xu Chen
中科院分区：
文献类型：
--
作者：
Shaohui Lin;Xiaoxi Zhang;Yupeng Li;Carlee Joe-Wong;Jingpu Duan;Xu Chen
深度学习（DL）驱动的实时应用程序通常需要使用地理位置生成的数据流进行持续训练。通过模型训练实现计算节点之间的数据卸载有望缓解生成大型数据集的设备可能具有低计算能力的问题。然而，卸载可能会损害模型收敛并产生通信成本，这必须与计算和模型同步所花费的成本相平衡。因此，本文提出了EdgeC3，这是一种新的框架，可以优化模型聚合的频率和连续生成的数据流的动态卸载，在长期准确性和成本之间进行权衡。我们首先提供了一个新的误差范围，以捕捉随着时间的推移和异构设备的数据动态的影响。在此基础上，我们设计了一个两时标的在线优化框架。我们周期性地学习同步频率，以适应不确定的未来卸载和网络变化。在更精细的时间尺度上，我们通过扩展李雅普诺夫优化技术来管理在线卸载，以处理非常规设置，在这种情况下，我们的长期全局约束可以突然改变在较长时间尺度上决定的聚合频率。最后，我们从理论上证明了EdgeC3的收敛性，通过整合我们的两个时间尺度的决策的耦合效应，我们通过大量的实验证明了它的优势。
Deep learning (DL) powered real-time applications usually need continuous training using data streams generated geographically. Enabling data offloading among computation nodes through model training is promising to mitigate the problem that devices generating large datasets may have low computation capability. However, offloading can compromise model convergence and incur communication costs, which must be balanced with the cost spent on computation and model synchronization. Therefore, this paper proposes EdgeC3, a novel framework that can optimize the frequency of model aggregation and dynamic offloading for continuously generated data streams, navigating the trade-off between long-term accuracy and cost. We first provide a new error bound to capture the impacts of data dynamics that are varying over time and heterogeneous across devices. Based on the bound, we design a two-timescale online optimization framework. We periodically learn the synchronization frequency to adapt with uncertain future offloading and network changes. In the finer timescale, we manage online offloading by extending Lyapunov optimization techniques to handle an unconventional setting, where our long-term global constraint can have abruptly changed aggregation frequencies that are decided in the longer timescale. Finally, we theoretically prove the convergence of EdgeC3 by integrating the coupled effects of our two-timescale decisions, and we demonstrate its advantage through extensive experiments.