Enhancing Federated Class-Incremental Learning via Spatial-Temporal Statistics Aggregation

TL;DR

This paper introduces a novel Spatial-Temporal Statistics Aggregation (STSA) framework for federated class-incremental learning, effectively addressing data heterogeneity and reducing communication overhead while improving performance.

Contribution

The paper proposes STSA, a unified spatial-temporal feature statistics aggregation method, and STSA-E, a communication-efficient variant with theoretical guarantees, advancing FCIL capabilities.

Findings

Outperforms state-of-the-art FCIL methods in accuracy.

Reduces communication and computation costs significantly.

Maintains robustness against data heterogeneity.

Abstract

Federated Class-Incremental Learning (FCIL) enables Class-Incremental Learning (CIL) from distributed data. Existing FCIL methods typically integrate old knowledge preservation into local client training. However, these methods cannot avoid spatial-temporal client drift caused by data heterogeneity and often incur significant computational and communication overhead, limiting practical deployment. To address these challenges simultaneously, we propose a novel approach, Spatial-Temporal Statistics Aggregation (STSA), which provides a unified framework to aggregate feature statistics both spatially (across clients) and temporally (across stages). The aggregated feature statistics are unaffected by data heterogeneity and can be used to update the classifier in closed form at each stage. Additionally, we introduce STSA-E, a communication-efficient variant with theoretical guarantees, achieving similar performance to STSA-E with much lower communication overhead. Extensive experiments on three widely used FCIL datasets, with varying degrees of data heterogeneity, show that our method outperforms state-of-the-art FCIL methods in terms of performance, flexibility, and both communication and computation efficiency. The code is available at https://github.com/Yuqin-G/STSA.