Li Dun, Han Dezhi, Weng Tien-Hsiung, Zheng Zibin, Li Hongzhi, Liu Han, Castiglione Arcangelo, Li Kuan-Ching
College of Information Engineering at Shanghai Maritime University, Pudong, China.
Department of Computer Science and Information Engineering, Providence University, Taichung City, Taiwan.
Soft comput. 2022;26(9):4423-4440. doi: 10.1007/s00500-021-06496-5. Epub 2021 Nov 20.
Federated learning () is a promising decentralized deep learning technology, which allows users to update models cooperatively without sharing their data. is reshaping existing industry paradigms for mathematical modeling and analysis, enabling an increasing number of industries to build privacy-preserving, secure distributed machine learning models. However, the inherent characteristics of have led to problems such as privacy protection, communication cost, systems heterogeneity, and unreliability model upload in actual operation. Interestingly, the integration with Blockchain technology provides an opportunity to further improve the security and performance, besides increasing its scope of applications. Therefore, we denote this integration of Blockchain and as the Blockchain-based federated learning () framework. This paper introduces an in-depth survey of and discusses the insights of such a new paradigm. In particular, we first briefly introduce the technology and discuss the challenges faced by such technology. Then, we summarize the Blockchain ecosystem. Next, we highlight the structural design and platform of . Furthermore, we present the attempts ins improving performance with Blockchain and several combined applications of incentive mechanisms in . Finally, we summarize the industrial application scenarios of .
联邦学习（）是一种很有前途的去中心化深度学习技术，它允许用户在不共享数据的情况下协同更新模型。正在重塑数学建模和分析的现有行业范式，使越来越多的行业能够构建隐私保护、安全的分布式机器学习模型。然而，的固有特性在实际操作中导致了隐私保护、通信成本、系统异构性和模型上传不可靠等问题。有趣的是，与区块链技术的集成除了扩大其应用范围外，还提供了进一步提高安全性和性能的机会。因此，我们将这种区块链与的集成称为基于区块链的联邦学习（）框架。本文对进行了深入调研，并讨论了这种新范式的见解。具体而言，我们首先简要介绍技术，并讨论该技术面临的挑战。然后，我们总结区块链生态系统。接下来，我们重点介绍的结构设计和平台。此外，我们展示了用区块链提高性能的尝试以及激励机制在中的几种联合应用。最后，我们总结的工业应用场景。
