Enhancing Local Model Diversity to Improve One-Shot Sequential Federated Learning for Non-IID Data
A novel one-shot sequential federated learning framework that enhances local model diversity to improve global model performance while reducing communication costs and mitigating the impact of non-IID data.