This study develops a federated learning (FL) framework overcoming largely incremental communication costs due to model sizes in typical frameworks without compromising model performance. To this end, based on the idea of leveraging an unlabeled open dataset, we propose a distillation-based semi-supervised FL (DS-FL) algorithm that exchanges the outputs of local models among mobile devices, instead of model parameter exchange employed by the typical frameworks. In DS-FL, the communication cost depends only on the output dimensions of the models and does not scale up according to the model size. The exchanged model outputs are used to label each sample of the open dataset, which creates an additionally labeled dataset. Based on the new dataset, local models are further trained, and model performance is enhanced owing to the data augmentation effect. We further highlight that in DS-FL, the heterogeneity of the devices' dataset leads to ambiguous of each data sample and lowing of the training convergence. To prevent this, we propose entropy reduction averaging, where the aggregated model outputs are intentionally sharpened. Moreover, extensive experiments show that DS-FL reduces communication costs up to 99% relative to those of the FL benchmark while achieving similar or higher classification accuracy.
翻译:这项研究开发了一个联合学习(FL)框架,由于典型框架中的模型规模大小而没有损及模型性能,在很大程度上增加了通信成本。为此,根据利用未贴标签的开放数据集的设想,我们提议采用基于蒸馏的半监督FL(DS-FL)算法,在移动设备之间交换当地模型的输出,而不是典型框架采用的模型参数交换。在DS-FL中,通信成本仅取决于模型的产出层面,不根据模型大小扩大规模。交换的模型产出被用于给开放数据集的每个样本贴上标签,从而产生额外的标签数据集。基于新的数据集,进一步培训当地模型,由于数据增强效应,模型性能得到提高。我们进一步强调,在DS-FL中,设备数据集的异质性导致每个数据样本的模糊性,降低培训趋同程度。为了防止这一点,我们提议降低英特罗比值的平均值,因为综合模型产出是有意精细化的。此外,广泛的实验显示,在新的数据集的基础上,对本地模型进行了进一步培训,由于数据增强作用,而使DS-FL的精确度降低了通信成本到类似的水平。