EDEN: Communication-Efficient and Robust Distributed Mean Estimation for Federated Learning
This addresses communication efficiency and robustness in federated learning, particularly for clients with diverse network conditions, but it appears incremental as it builds on existing DME methods.
The paper tackles the problem of inaccurate gradient estimation in federated learning due to communication constraints and network issues like packet losses, proposing EDEN, which shows consistent improvements over state-of-the-art techniques in distributed mean estimation.
Distributed Mean Estimation (DME) is a central building block in federated learning, where clients send local gradients to a parameter server for averaging and updating the model. Due to communication constraints, clients often use lossy compression techniques to compress the gradients, resulting in estimation inaccuracies. DME is more challenging when clients have diverse network conditions, such as constrained communication budgets and packet losses. In such settings, DME techniques often incur a significant increase in the estimation error leading to degraded learning performance. In this work, we propose a robust DME technique named EDEN that naturally handles heterogeneous communication budgets and packet losses. We derive appealing theoretical guarantees for EDEN and evaluate it empirically. Our results demonstrate that EDEN consistently improves over state-of-the-art DME techniques.