Class-wise Balancing Data Replay for Federated Class-Incremental Learning

TL;DR

This paper introduces FedCBDR, a novel federated class-incremental learning method that uses class-wise balancing data replay and adaptive temperature scaling to mitigate class imbalance and improve model accuracy across tasks.

Contribution

FedCBDR is the first to combine class-wise balanced replay with adaptive temperature scaling for federated class-incremental learning, addressing class imbalance issues effectively.

Findings

Achieves 2%-15% higher Top-1 accuracy than state-of-the-art methods.

Balances class-wise sampling under heterogeneous data distributions.

Improves generalization across tasks with different data imbalances.

Abstract

Federated Class Incremental Learning (FCIL) aims to collaboratively process continuously increasing incoming tasks across multiple clients. Among various approaches, data replay has become a promising solution, which can alleviate forgetting by reintroducing representative samples from previous tasks. However, their performance is typically limited by class imbalance, both within the replay buffer due to limited global awareness and between replayed and newly arrived classes. To address this issue, we propose a class wise balancing data replay method for FCIL (FedCBDR), which employs a global coordination mechanism for class-level memory construction and reweights the learning objective to alleviate the aforementioned imbalances. Specifically, FedCBDR has two key components: 1) the global-perspective data replay module reconstructs global representations of prior task in a privacy-preserving manner, which then guides a class-aware and importance-sensitive sampling strategy to achieve balanced replay; 2) Subsequently, to handle class imbalance across tasks, the task aware temperature scaling module adaptively adjusts the temperature of logits at both class and instance levels based on task dynamics, which reduces the model's overconfidence in majority classes while enhancing its sensitivity to minority classes. Experimental results verified that FedCBDR achieves balanced class-wise sampling under heterogeneous data distributions and improves generalization under task imbalance between earlier and recent tasks, yielding a 2%-15% Top-1 accuracy improvement over six state-of-the-art methods.