An Contract Theory based Federated Learning Aggregation Algorithm in IoT Network
Mengmeng Tian · Journal of Physics Conference Series · 2022
Abstract Taking IoT devices as the edge nodes is one of the powerful way to offloading the federated task since IoT devices are closer to the data generation end. The aggregation efficiency of federated learning in the IoT environment is inefficiency since the server of federated learning can not know the data quality of heterogeneous IoT device. How to encourage IoT edge clients to participate in federated learning and maximize the aggregation effect of the global model is an important problem. This paper proposes a federated learning aggregation model based on contract theory incentive mechanism. Our experimental results show that the proposed algorithm effectively improves the aggregation efficiency of federated learning compared with FedAvg algorithm.