【Author】
Awan, Sana; Li, Fengjun; Luo, Bo; Liu, Mei
【Source】PROCEEDINGS OF THE 2019 ACM SIGSAC CONFERENCE ON COMPUTER AND COMMUNICATIONS SECURITY (CCS'19)
【Abstract】Federated learning (FL) is promising in supporting collaborative learning applications that involve large datasets, massively distributed data owners and unreliable network connectivity. To protect data privacy, existing FL approaches adopt (k, n)-threshold secret sharing schemes, based on the semi-honest assumption for clients, to enable secure multiparty computation in local model update exchange which deals with random client dropouts at the cost of increasing data size. These approaches adopt the semi-honest assumption for clients, therefore they are vulnerable to malicious clients. In this work, we propose a blockchain-based privacy-preserving federated learning (BC-based PPFL) framework, which leverages the immutability and decentralized trust properties of blockchain to provide provenance of model updates. Our proof-of-concept implementation of BC-based PPFL demonstrates it is practical for secure aggregation of local model updates in the federated setting.
【Keywords】Federated Learning; Privacy; Blockchain
【标题】Poster：使用区块链的可靠且负责任的隐私保护联邦学习框架
【摘要】联邦学习 (FL) 在支持涉及大型数据集、大规模分布式数据所有者和不可靠网络连接的协作学习应用程序方面很有前景。为了保护数据隐私，现有的 FL 方法采用 (k, n) 阈值秘密共享方案，基于对客户端的半诚实假设，以在本地模型更新交换中实现安全的多方计算，以处理随机客户端丢失为代价增加数据量。这些方法对客户端采用半诚实的假设，因此它们容易受到恶意客户端的攻击。在这项工作中，我们提出了一个基于区块链的隐私保护联邦学习（基于 BC 的 PPFL）框架，该框架利用区块链的不变性和去中心化信任属性来提供模型更新的来源。我们基于 BC 的 PPFL 的概念验证实现表明，它对于在联合设置中安全聚合本地模型更新是实用的。
评论