Enhancing Privacy in Federated Learning through Local Training
This work addresses privacy and efficiency issues in federated learning for distributed computing systems, but it appears incremental as it builds on existing local training methods.
The paper tackles the challenges of expensive communications and privacy preservation in federated learning by proposing Fed-PLT, which uses local training to reduce communication rounds without impacting accuracy, and derives differential privacy bounds dependent on local training epochs.
In this paper we propose the federated learning algorithm Fed-PLT to overcome the challenges of (i) expensive communications and (ii) privacy preservation. We address (i) by allowing for both partial participation and local training, which significantly reduce the number of communication rounds between the central coordinator and computing agents. The algorithm matches the state of the art in the sense that the use of local training demonstrably does not impact accuracy. Additionally, agents have the flexibility to choose from various local training solvers, such as (stochastic) gradient descent and accelerated gradient descent. Further, we investigate how employing local training can enhance privacy, addressing point (ii). In particular, we derive differential privacy bounds and highlight their dependence on the number of local training epochs. We assess the effectiveness of the proposed algorithm by comparing it to alternative techniques, considering both theoretical analysis and numerical results from a classification task.