Blockchain-based Secure Aggregation for Federated Learning with a Traffic Prediction Use Case
Qiong Zhang, Paparao Palacharla, Motoyoshi Sekiya, Junichi Suga, Toru Katagiri · 2021
Federated learning is a distributed machine learning approach that can be applied to many networking applications. In this paper, we propose a novel blockchain-based secure aggregation protocol for federated learning, which simplifies the existing secure aggregation process by leveraging consensus through blockchain. We demonstrate the prototype by training a general LSTM model for traffic prediction at cell sites based on distributed time series datasets.