Helios:具有动态平衡协作的异构感知联邦学习
Helios: Heterogeneity-Aware Federated Learning with Dynamically Balanced Collaboration
摘要:
在本文中,我们提出了太阳神,异质性意识FL框架来解决掉队的问题。Helios识别各个设备的异构训练能力,因此识别关于协作训练速度的预期神经网络模型训练量。对于离散设备,提出了一种“软训练”方法,通过旋转神经元训练方法将原始相同的训练模型动态压缩到期望的体积。通过广泛的算法分析和优化方案,可以加速落后者,同时保持局部训练和联邦协作的收敛性。
英文摘要:
In this paper, we propose Helios, a heterogeneity-aware FL framework to tackle the straggler issue. Helios identifies individual devices' heterogeneous training capability, and therefore the expected neural network model training volumes regarding the collaborative training pace. For straggling devices, a "soft-training" method is proposed to dynamically compress the original identical training model into the expected volume through a rotating neuron training approach. With extensive algorithm analysis and optimization schemes, the stragglers can be accelerated while retaining the convergence for local training as well as federated collaboration.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。