MOHAWK: Mobility and Heterogeneity-Aware Dynamic Community Selection for Hierarchical Federated Learning
MOHAWK: Mobility and Heterogeneity-Aware Dynamic Community Selection for Hierarchical Federated Learning
复制标题
DOI：
10.1145/3576842.3582378
复制
发表时间：
2023-05
期刊：
影响因子：
--
通讯作者：
Allen-Jasmin Farcas;Myungjin Lee;R. Kompella;Hugo Latapie;G. de Veciana;R. Marculescu
中科院分区：
文献类型：
--
作者：
Allen-Jasmin Farcas;Myungjin Lee;R. Kompella;Hugo Latapie;G. de Veciana;R. Marculescu
联邦学习（FL）的最新发展集中在优化数据，硬件和模型异质性的学习过程。然而，大多数方法都假设所有设备都是静止的，正在充电，并且在本地数据上训练时始终连接到Wi-Fi。我们认为，当真实的设备四处移动时，FL过程受到负面影响，并且用于通信的设备能量增加。为了减轻这种影响，我们提出了一个动态的社区选择算法，提高了通信的能源效率和两个新的聚合策略，提高学习性能的分层FL（HFL）。对于真实的移动轨迹，我们表明，与最先进的HFL解决方案相比，我们的方法是可扩展的，在多个数据集上实现了更好的准确性，收敛速度高达3.88倍，并且对于IID和非IID机器人都具有更高的能效。
The recent developments in Federated Learning (FL) focus on optimizing the learning process for data, hardware, and model heterogeneity. However, most approaches assume all devices are stationary, charging, and always connected to the Wi-Fi when training on local data. We argue that when real devices move around, the FL process is negatively impacted and the device energy spent for communication is increased. To mitigate such effects, we propose a dynamic community selection algorithm which improves the communication energy efficiency and two new aggregation strategies that boost the learning performance in Hierarchical FL (HFL). For real mobility traces, we show that compared to state-of-the-art HFL solutions, our approach is scalable, achieves better accuracy on multiple datasets, converges up to 3.88 × faster, and is significantly more energy efficient for both IID and non-IID scenarios.1