Communication Efficient Private Federated Learning Using Dithering
This addresses privacy and communication efficiency in federated learning, but it is incremental as it builds on existing trusted aggregator models and quantization techniques.
The paper tackles the challenge of preserving privacy and ensuring efficient communication in federated learning by proposing a quantization scheme based on subtractive dithering at clients, which replicates normal noise addition at the aggregator to guarantee the same differential privacy level while substantially reducing communication compared to transmitting full precision gradients.
The task of preserving privacy while ensuring efficient communication is a fundamental challenge in federated learning. In this work, we tackle this challenge in the trusted aggregator model, and propose a solution that achieves both objectives simultaneously. We show that employing a quantization scheme based on subtractive dithering at the clients can effectively replicate the normal noise addition process at the aggregator. This implies that we can guarantee the same level of differential privacy against other clients while substantially reducing the amount of communication required, as opposed to transmitting full precision gradients and using central noise addition. We also experimentally demonstrate that the accuracy of our proposed approach matches that of the full precision gradient method.