Tackling Heterogeneity in Medical Federated learning via Vision Transformers
This addresses the problem of data imbalance in medical federated learning for healthcare applications, offering a method that is incremental over existing regularization approaches.
The paper tackled data heterogeneity in medical federated learning by using Vision Transformers, which improved performance for underrepresented clients without significantly reducing overall accuracy.
Optimization-based regularization methods have been effective in addressing the challenges posed by data heterogeneity in medical federated learning, particularly in improving the performance of underrepresented clients. However, these methods often lead to lower overall model accuracy and slower convergence rates. In this paper, we demonstrate that using Vision Transformers can substantially improve the performance of underrepresented clients without a significant trade-off in overall accuracy. This improvement is attributed to the Vision transformer's ability to capture long-range dependencies within the input data.