Federated learning keeps data private while reducing communication
Privacy-enhanced federated learning via asynchronous aggregation and local differential perturbation
Machine LearningArtificial IntelligenceComputational Engineering, Finance, and ScienceDatabases
Summary
Training machine learning models across multiple devices usually risks revealing private data. The authors propose a new method that protects privacy by mixing advanced mathematical tricks and encryptions during training. Their technique allows devices to share model updates without exposing sensitive information and works well even when updates happen at different times. Tests show the method keeps model accuracy high while lowering the amount of data sent between devices.
What this means in practice
- •For enterprise data teams: Train machine learning models collaboratively across multiple offices while ensuring sensitive data stays private and reducing network load.
- •For mobile app developers: Build apps that learn from user data on device securely and efficiently, preserving privacy even with irregular data syncing.
Authors
Zhen Zhong, Shini Yang, Liesheng Wei
Abstract
This study proposes a privacy-enhanced federated learning framework to address secure collaborative training in distributed data environments. The framework integrates Dynamic Differential Privacy (DDP), lightweight Homomorphic Encryption (HE), and Local Differential Privacy (LDP) mechanisms to ensure data privacy protection during model training. Additionally, the framework employs an asynchronous aggregation strategy with version control to support distributed training in asynchronous environments. Experimental validation on the CIFAR-10 and Purchase-100 benchmark datasets demonstrates that the method maintains high classification accuracy (up to 82.6%) even under stringent privacy constraints (ε = 0.1), while reducing communication overhead by 21.3% compared to FedAvg. Experimental results demonstrate that this framework effectively balances privacy protection and model performance in distributed machine learning scenarios, providing a scalable technical foundation for large-scale distributed collaborative computing.