IEEE

Exploring the Privacy-Accuracy Trade-Off Using Adaptive Gradient Clipping in Federated Learning

Abstract: In Differentially Private Federated Learning (DP-FL), gradient clipping can prevent excessive noise from being added to the gradient and ensure that the impact of noise is within a ...
IEEE

Using Highly Compressed Gradients in Federated Learning for Data Reconstruction Attacks

Abstract: Federated learning (FL) preserves data privacy by exchanging gradients instead of local training data. However, these private data can still be reconstructed from the exchanged gradients.