Optimize Scheduling of Federated Learning on Battery-powered Mobile Devices
Optimize Scheduling of Federated Learning on Battery-powered Mobile Devices
复制标题
DOI：
10.1109/ipdps47924.2020.00031
复制
发表时间：
2020-05
期刊：
影响因子：
--
通讯作者：
Cong Wang;Xin Wei;Pengzhan Zhou
中科院分区：
文献类型：
--
作者：
Cong Wang;Xin Wei;Pengzhan Zhou
联合学习通过从移动设备中汇总本地计算的更新来了解隐私的过程，而当前的研究通常会优先考虑交流间接费用，从而证明了一项经验研究此外，如果课程在移动设备之间是不平衡的差异和准确性损失在本文中，我们将数据作为可调旋钮来安排训练，并基于离线分析实现近乎最佳的计算解决方案。是类平衡或不平衡的。通过可忽略的精度损失达到210倍的加速。
Federated learning learns a collaborative model by aggregating locally-computed updates from mobile devices for privacy preservation. While current research typically prioritizing the minimization of communication overhead, we demonstrate from an empirical study, that computation heterogeneity is a more pronounced bottleneck on battery-powered mobile devices. Moreover, if class is unbalanced among the mobile devices, inappropriate selection of participants may adversely cause gradient divergence and accuracy loss. In this paper, we utilize data as a tunable knob to schedule training and achieve near-optimal solutions of computation time and accuracy loss. Based on the offline profiling, we formulate optimization problems and propose polynomial-time algorithms when data is class-balanced or unbalanced. We evaluate the optimization framework extensively on a mobile testbed with two datasets. Compared with common benchmarks of federated learning, our algorithms achieve 210× speedups with negligible accuracy loss. They also mitigate the impact from mobile stragglers and improve parallelism for federated learning.