With the help of a new architecture called Edge/Fog (E/F) computing, cloud computing services can now be extended nearer to data generator devices. E/F computing in combination with Deep Learning (DL) is a promisedtechnique that is vastly applied in numerous fields. To train their models, data producers in conventional DL architectures with E/F computing enable them to repeatedly transmit and communicate data with third-party servers, like Edge/Fog or cloud servers. Due to the extensive bandwidth needs, legal issues, and privacy risks, this architecture is frequently impractical. Through a centralized server, the models can be co-trained by FL through distributed clients, including cars, hospitals, and mobile phones, while preserving data localization. As it facilitates group learning and model optimization, FL can therefore be seen as a motivating element in the E/F computing paradigm. Although FL applications in E/F computing environments have been considered in previous studies, FL execution and hurdles in the E/F computing framework have not been thoroughly covered. In order to identify advanced solutions, this chapter will provide a review of the application of FL in E/F computing systems. We think that by doing this chapter, researchers will learn more about how E/F computing and FL enable related concepts and technologies. Some case studies about the implementation of federated learning in E/F computing are being investigated. The open issues and future research directions are introduced.

翻译：随着名为边缘/雾（E/F）计算的新型架构的问世，云计算服务现已能够扩展至更靠近数据生成设备的位置。E/F计算与深度学习（DL）相结合，是一项被广泛应用于众多领域的极具前景的技术。在传统的基于E/F计算的DL架构中，数据生产者为了训练模型，需要反复将数据传输至第三方服务器（如边缘/雾或云服务器）并与之通信。由于对带宽需求巨大，且存在法律问题与隐私风险，这种架构通常并不可行。通过一个中心化服务器，联邦学习（FL）能利用分布式客户端（包括汽车、医院和手机）在不移动数据的情况下协同训练模型。由于FL促进了群体学习和模型优化，因此可被视为E/F计算范式中的一个激励性要素。尽管先前的研究已考虑了FL在E/F计算环境中的应用，但关于FL在E/F计算框架中的执行过程及所遇障碍的探讨尚不充分。为了识别先进的解决方案，本章将综述FL在E/F计算系统中的应用。我们相信，通过本章的论述，研究人员将更深入地了解E/F计算与FL如何赋能相关概念与技术。本章还探究了若干关于在E/F计算中实施联邦学习的案例研究，并介绍了当前的开放性问题和未来的研究方向。