Fine-tuning Global Model via Data-Free Knowledge Distillation for Non-IID Federated Learning
This addresses performance drops in federated learning due to non-IID data, offering a plugin solution for existing methods, though it is incremental as it builds on prior FL techniques.
The paper tackles the challenge of data heterogeneity in federated learning, which causes slow convergence and degraded performance, by proposing a data-free knowledge distillation method to fine-tune the global model, resulting in significant outperformance over state-of-the-art FL algorithms.
Federated Learning (FL) is an emerging distributed learning paradigm under privacy constraint. Data heterogeneity is one of the main challenges in FL, which results in slow convergence and degraded performance. Most existing approaches only tackle the heterogeneity challenge by restricting the local model update in client, ignoring the performance drop caused by direct global model aggregation. Instead, we propose a data-free knowledge distillation method to fine-tune the global model in the server (FedFTG), which relieves the issue of direct model aggregation. Concretely, FedFTG explores the input space of local models through a generator, and uses it to transfer the knowledge from local models to the global model. Besides, we propose a hard sample mining scheme to achieve effective knowledge distillation throughout the training. In addition, we develop customized label sampling and class-level ensemble to derive maximum utilization of knowledge, which implicitly mitigates the distribution discrepancy across clients. Extensive experiments show that our FedFTG significantly outperforms the state-of-the-art (SOTA) FL algorithms and can serve as a strong plugin for enhancing FedAvg, FedProx, FedDyn, and SCAFFOLD.