Preserving Privacy in Federated Learning with Ensemble Cross-Domain Knowledge Distillation
This work addresses privacy and efficiency issues in federated learning, which is crucial for applications like healthcare and finance, though it appears incremental as it builds on existing FL and knowledge distillation techniques.
The paper tackled the problems of privacy leakage and communication bottlenecks in federated learning by developing a method using one-shot offline knowledge distillation with cross-domain public data, achieving superior accuracy and communication efficiency in image and text classification tasks.
Federated Learning (FL) is a machine learning paradigm where local nodes collaboratively train a central model while the training data remains decentralized. Existing FL methods typically share model parameters or employ co-distillation to address the issue of unbalanced data distribution. However, they suffer from communication bottlenecks. More importantly, they risk privacy leakage. In this work, we develop a privacy preserving and communication efficient method in a FL framework with one-shot offline knowledge distillation using unlabeled, cross-domain public data. We propose a quantized and noisy ensemble of local predictions from completely trained local models for stronger privacy guarantees without sacrificing accuracy. Based on extensive experiments on image classification and text classification tasks, we show that our privacy-preserving method outperforms baseline FL algorithms with superior performance in both accuracy and communication efficiency.