FedKPer: Tackling Generalization and Personalization in Medical Federated Learning via Knowledge Personalization

2026-05-01

AI Summary

This work addresses the challenges of data heterogeneity in medical federated learning, which often leads to poor global model generalization, insufficient local personalization, and catastrophic forgetting. To tackle these issues jointly, the study introduces knowledge personalization directly into the federated training process: personalized models are obtained via knowledge distillation during local updates, while a novel aggregation scheme incorporates client reliability and label diversity through adaptive weighting, complemented by selective model alignment. This unified approach simultaneously optimizes generalization, personalization, and mitigation of catastrophic forgetting. Extensive experiments in medical scenarios demonstrate that the proposed method achieves a superior trade-off among these competing objectives, significantly outperforming existing state-of-the-art approaches.

Abstract

Federated learning (FL) holds great potential for medical applications. However, statistical heterogeneity across healthcare institutions poses a major challenge for FL, as the global model struggles both to generalize across unseen patient populations and to adapt to the unique data distributions of individual hospitals. This heterogeneity also exacerbates forgetting at both the global and local level, resulting in previous learned patient patterns to be misclassified after model updates. While prior work has largely treated generalization and personalization as separate challenges, we show that a better balance between the two can be achieved through selective alignment with the global model and a modified aggregation scheme, which together mitigate the effects of statistical heterogeneity. Specifically, we introduce FedKPer, which introduces knowledge personalization into the training stage of each local device. Afterwards, generalization is considered via the global model aggregation process, where local updates that are reliable and label-diverse are emphasized. We evaluate the performance of FedKPer, devising additional metrics that relate to common consequences of forgetting. Overall, we demonstrate FedKPer improves the generalization-personalization trade-off without sacrificing retention.

Problem

• federated learning
• statistical heterogeneity
• generalization
• personalization
• catastrophic forgetting

Innovation

• federated learning
• knowledge personalization
• statistical heterogeneity
• generalization-personalization trade-off
• catastrophic forgetting