Federated Learning via Synthetic Data

Abstract

Federated learning allows for the training of a model using data on multiple clients without the clients transmitting that raw data. However the standard method is to transmit model parameters (or updates), which for modern neural networks can be on the scale of millions of parameters, inflicting significant computational costs on the clients. We propose a method for federated learning where instead of transmitting a gradient update back to the server, we instead transmit a small amount of synthetic `data'. We describe the procedure and show some experimental results suggesting this procedure has potential, providing more than an order of magnitude reduction in communication costs with minimal model degradation.

References in corpus (2)

• Communication-Efficient Learning of Deep Networks from Decentralized Data
• Soft-Label Dataset Distillation and Text Dataset Distillation

Cited by in corpus (5)

• A Comprehensive Survey of Dataset Distillation
• Dataset Condensation with Gradient Matching
• No Fear of Heterogeneity: Classifier Calibration for Federated Learning with Non-IID Data
• Towards Fair Federated Learning with Zero-Shot Data Augmentation
• Synthetic data shuffling accelerates the convergence of federated learning under data heterogeneity