The ubiquity of camera-enabled devices has led to large amounts of unlabeled image data being produced at the edge. The integration of self-supervised learning (SSL) and federated learning (FL) into one coherent system can potentially offer data privacy guarantees while also advancing the quality and robustness of the learned visual representations without needing to move data around. However, client bias and divergence during FL aggregation caused by data heterogeneity limits the performance of learned visual representations on downstream tasks. In this paper, we propose a new aggregation strategy termed Layer-wise Divergence Aware Weight Aggregation (L-DAWA) to mitigate the influence of client bias and divergence during FL aggregation. The proposed method aggregates weights at the layer-level according to the measure of angular divergence between the clients' model and the global model. Extensive experiments with cross-silo and cross-device settings on CIFAR-10/100 and Tiny ImageNet datasets demonstrate that our methods are effective and obtain new SOTA performance on both contrastive and non-contrastive SSL approaches.

翻译：配备摄像头的设备无处不在，导致边缘端产生了大量无标注图像数据。将自监督学习（SSL）与联邦学习（FL）整合到一个连贯系统中，既能在无需移动数据的情况下保障数据隐私，又能提升所学视觉表示的质量和鲁棒性。然而，FL聚合过程中由数据异质性导致的客户端偏差和散度限制了所学视觉表示在下游任务上的性能。本文提出一种名为层级发散感知权重聚合（L-DAWA）的新聚合策略，以减轻FL聚合中客户端偏差和散度的影响。该方法根据客户端模型与全局模型之间的角度散度度量，在层级上进行权重聚合。在CIFAR-10/100和Tiny ImageNet数据集上进行的跨孤岛和跨设备场景大量实验表明，我们的方法效果显著，并在对比式和非对比式SSL方法上均取得了新的最佳性能。