Record and Reward Federated Learning Contributions with Blockchain

【Author】 Martinez, Ismael; Francis, Sreya; Hafid, Abdelhakim Senhaji

【Source】2019 INTERNATIONAL CONFERENCE ON CYBER-ENABLED DISTRIBUTED COMPUTING AND KNOWLEDGE DISCOVERY (CYBERC)

【Abstract】Although Federated Learning allows for participants to contribute their local data without it being revealed, it faces issues in data security and in accurately paying participants for quality data contributions. In this paper, we propose an EOS Blockchain design and workflow to establish data security, a novel validation error based metric upon which we qualify gradient uploads for payment, and implement a small example of our blockchain Federated Learning model to analyze its performance.

【Keywords】blockchain; Federated Learning; distributed machine learning; class sampled validation error

【标题】使用区块链记录和奖励联邦学习贡献

【摘要】尽管联邦学习允许参与者贡献他们的本地数据而不被泄露，但它面临着数据安全和准确支付参与者的质量数据贡献的问题。在本文中，我们提出了一种 EOS 区块链设计和工作流程来建立数据安全性，这是一种基于验证错误的新指标，我们可以根据该指标对梯度上传进行支付，并实施我们的区块链联邦学习模型的一个小示例来分析其性能。

【关键词】区块链；联邦学习；分布式机器学习；类抽样验证错误

【发表时间】2019

【收录时间】2022-07-06

【文献类型】Proceedings Paper

【论文大主题】区块链联邦学习

【论文小主题】联邦学习为主体

【翻译者】石东瑛