FeDMRA: Federated Incremental Learning with Dynamic Memory Replay Allocation

TL;DR

This paper introduces FeDMRA, a federated incremental learning method with dynamic memory allocation that improves model performance in non-IID healthcare data scenarios by balancing client fairness and mitigating forgetting.

Contribution

It proposes a novel dynamic memory allocation strategy for federated class-incremental learning that leverages data heterogeneity and enhances fairness among clients.

Findings

Significant performance improvements over baseline models on medical image datasets.

Effective mitigation of catastrophic forgetting in non-IID federated settings.

Balanced resource allocation among clients enhances overall model accuracy.

Abstract

In federated healthcare systems, Federated Class-Incremental Learning (FCIL) has emerged as a key paradigm, enabling continuous adaptive model learning among distributed clients while safeguarding data privacy. However, in practical applications, data across agent nodes within the distributed framework often exhibits non-independent and identically distributed (non-IID) characteristics, rendering traditional continual learning methods inapplicable. To address these challenges, this paper covers more comprehensive incremental task scenarios and proposes a dynamic memory allocation strategy for exemplar storage based on the data replay mechanism. This strategy fully taps into the inherent potential of data heterogeneity, while taking into account the performance fairness of all participating clients, thereby establishing a balanced and adaptive solution to mitigate catastrophic forgetting. Unlike the fixed allocation of client exemplar memory, the proposed scheme emphasizes the rational allocation of limited storage resources among clients to improve model performance. Furthermore, extensive experiments are conducted on three medical image datasets, and the results demonstrate significant performance improvements compared to existing baseline models.