Decentralized Deep Learning for Multi-Access Edge Computing: A Survey on Communication Efficiency and Trustworthiness

Wider coverage and a better solution to latency reduction in 5G necessitate its combination with multi-access edge computing technology. Decentralized deep learning (DDL), such as federated swarm learning, promising privacy-preserving data processing for millions of smart devices leverages distributed multilayer neural networks within the networking local clients, without disclosing original training data. Notably, industries finance healthcare, where sensitive transactions personal medical records are cautiously maintained, DDL can facilitate collaboration among these institutes improve performance trained models while protecting privacy participating clients. In this survey article, we demonstrate technical fundamentals that benefit many walks society through decentralized learning. Furthermore, offer comprehensive overview current state art field by outlining challenges most relevant solutions from novel perspectives communication efficiency trustworthiness.